[ron178]

Originally Posted by rpm

The suffix 2 is used for a 2nd car of the same iteration/make, on which a 2nd test of similar initial code (OD and MD in this case) is being conducted, provided that it arrived at the lab the same week as the first one.

Precisely.

Originally Posted by rpm

I agree with your assessment that the video shown for the side impact test, does not qualify as 'far-side' test because of the absence of dummies in seats adjacent to the side being struck by the MDB.

Actually, it does not qualify as far-side impact due to the presence of dummies on the side struck by the MDB trolley. If the passenger side was struck with the dummies still on the driver side, that would qualify as far-side impact.

Originally Posted by rpm

This is evident in their proactive approach of getting ESC tested on both, the Thar and XUV 300. They even sponsored two new XUVs for the safer choice award, so that pedestrian protection and ESC could be tested.

The Thar is actually an interesting case since Mahindra sponsored all tests required for the Safer Choice Award, though they didn't qualify in the ODB test, which leads me to believe they really were aiming for 5 stars. When they realised they wouldn't qualify, they must have sponsored all tests anyway since they prepared for them (though the ESC test did not go too well).

Now one may argue that the Thar's ODB test score is a fair bit lower than that required for 5 stars. Remember how I mentioned earlier that many 4-star cars lose out on the minimum 14/16 in the ODB test because of the knee modifiers? I have a feeling Mahindra planned to conduct a verification sled test for having the knee modifiers removed, but couldn't do so because Global NCAP will not accept knee mapping data if an unstable footwell modifier has been applied. Hence the cumulative effect of the driver knee modifiers and the unstable footwell modifier brought down the score a fair bit. Tata and Mahindra's latest tested vehicles, the Tigor and the Thar, both have structural modifiers applied and hence would not qualify for removal of knee modifiers (though Tata and Mahindra obviously know the process to have them removed). Just as well, since Global NCAP wouldn't want a car with an unstable structure/footwell scoring the 5th star.

Knee mapping data will be accepted under the conditions below:
- The driver and front passenger’s head, neck, chest score are orange, yellow or green.
- Femur loads <3.8kN in the full vehicle test.
- Knee Slider <6mm in the full vehicle test.
- No structural modifiers applied i.e. integrity of the passenger compartment and/or footwell rupture. - A-pillar displacements must be below 65mm (using the standard Euro NCAP measurement).

[Source - Page 8 3.2.13 - Removal of Knee Modifiers]

Originally Posted by rpm

Tata is willing to sponsor the minimum number of vehicles required for that particular rating, be it 4 or 5-star.

Keep in mind that the Nexon was not equipped with ESC at launch. Only recently was it made standard across the range.

Originally Posted by rpm

Mind you, this isn't a bad thing at all in a country where the manufacturer with the highest market share outright refuses to send their vehicles for a test and calls GNCAP a self proclaimed party, while the manufacturer which is second on the sales chart is better at knowing its mistakes and keeping quiet instead of making ridiculous accusations and citing vague statistics to justify their mistake.

A manufacturer in a developed nation wouldn't dare challenge the Euro NCAP because that would hurt their reputation. They simply won't take the risk because in a sea of five-star results, even 4 stars are disappointing (look at this example and this one). This is precisely why I was disappointed at the rumour that Škoda is targeting 4 stars for the Kushaq. While we're not quite there yet, sooner or later every manufacturer will have hopped on the bandwagon for safer cars.

Originally Posted by rpm

Mahindra, on the other hand, is going the extra mile, and sponsoring vehicles for additional tests which make no difference to their rating. Case in point: the side impact test and ESC test for Thar.

Very well, since additional tests couldn't hurt. If anything, Global NCAP could use the same test data to assign ratings under their new protocol next year without having to conduct tests again (Latin NCAP did it with the Ford Ka multiple times where they used old test data to assign new ratings). Sponsoring extra tests is completely worth the returns for the manufacturer in the long run. Plus, ESC testing is largely non-destructive (unless it goes wrong). The XUV300's ESC test was sponsored after its impact tests but if you look at the Thar (for which ESC was tested on two cars) it looks like the ESC test cars were reused for the impact tests.

I'm quite surprised that Tata hasn't sponsored an ESC test for the Nexon yet, even after they've made the system standard across all trims.

[rpm]

Originally Posted by ron178

Actually, it does not qualify as far-side impact due to the presence of dummies on the side struck by the MDB trolley. If the passenger side was struck with the dummies still on the driver side, that would qualify as far-side impact.

Thank you for pointing it out and correcting me. I somehow had this video below stuck in my head, and hence assumed that occupant to occupant interaction is always checked using both the passengers. As it turns out, this is not necessarily the case, unless a centre airbag is present.

https://youtu.be/GOk3Q4chh4g

Where vehicles are equipped with a counter-measure (for example a centre airbag), Euro NCAP will install two front seat occupants in either the full-scale side barrier or side pole tests to demonstrate that the proposed solution works robustly, that it is effective for impacts from both sides of the car, and that it mitigates interaction between the front seat occupants. Vehicles without a counter-measure will have only one dummy in the full-scale tests.

Source: https://www.euroncap.com/en/vehicle-...r-side-impact/

Now one may argue that the Thar's ODB test score is a fair bit lower than that required for 5 stars. Remember how I mentioned earlier that many 4-star cars lose out on the minimum 14/16 in the ODB test because of the knee modifiers? I have a feeling Mahindra planned to conduct a verification sled test for having the knee modifiers removed, but couldn't do so because Global NCAP will not accept knee mapping data if an unstable footwell modifier has been applied. Hence the cumulative effect of the driver knee modifiers and the unstable footwell modifier brought down the score a fair bit. Tata and Mahindra's latest tested vehicles, the Tigor and the Thar, both have structural modifiers applied and hence would not qualify for removal of knee modifiers (though Tata and Mahindra obviously know the process to have them removed). Just as well, since Global NCAP wouldn't want a car with an unstable structure/footwell scoring the 5th star.

This is a very sound argument against the assumptions in my earlier post. In fact, I agree with the possibility of such a scenario, now that you have reminded me about the possibility of removal of knee modifiers. No objections to this.

Keep in mind that the Nexon was not equipped with ESC at launch. Only recently was it made standard across the range.

This is something I had not considered while suggesting that Tata sends only the minimum required vehicles for a star rating. I was looking at the 4-star Tigor when I came to this conclusion. I guess I was more fixated on the fact that Tigor was not sponsored for MDB test, because it wasn't needed for a 4-star rating, while conveniently forgetting that the Thar is a one-off and Mahindra's very own Marazzo was not sponsored for MDB either.

A manufacturer in a developed nation wouldn't dare challenge the Euro NCAP because that would hurt their reputation. They simply won't take the risk because in a sea of five-star results, even 4 stars are disappointing (look at this example and this one). This is precisely why I was disappointed at the rumour that Škoda is targeting 4 stars for the Kushaq. While we're not quite there yet, sooner or later every manufacturer will have hopped on the bandwagon for safer cars.

You have perfectly echoed the sentiments of David Ward (President & CEO, GNCAP) with your first line, who says

Suzuki would never ever dare say something like that about the Japanese NCAP

Source:

https://youtu.be/tP8zwE7CY-U?t=1645

at 27:25

I have a different take about Skoda's approach though. For the developed nations, 4-star is a rarity in the sense that most results are 5-star, as you have rightly said and pointed out in your examples. However, in India, a 4-star is a rarity for the entirely opposite reason, so 4-star and above is currently a plus in my book, especially in context of the Indian market, because the Indian consumer (largely the casual buyer, which is the majority) isn't very concerned or even aware about crash ratings. This is from personal experience, be it friends, or family. Crash rating is simply not a priority for the market, so I'll take any 4-star and above result I can get, precisely because the market for that very rating is minuscule. In fact, I fear that if the government compliance norms do not evolve with time (to their credit, ABS & two airbags as standard is a very welcome move), we may have a scenario where the current buzz dies out, and manufacturers making affordable safe cars go out of business because prioritising safety clearly wasn't in line with the market demands of that time.

However, if and when 4-star becomes commonplace enough, I would love a push towards 5-star products. I can understand the disappointment though. Skoda being an international player, already has experience with even more stringent norms, yet isn't able to compete with homegrown products from a segment below— brands with significantly lesser international presence. But when I assess this in the context of the only tested rival being barely 3-star, I can't help but look on the bright side.

Very well, since additional tests couldn't hurt. If anything, Global NCAP could use the same test data to assign ratings under their new protocol next year without having to conduct tests again (Latin NCAP did it with the Ford Ka multiple times where they used old test data to assign new ratings). Sponsoring extra tests is completely worth the returns for the manufacturer in the long run. Plus, ESC testing is largely non-destructive (unless it goes wrong). The XUV300's ESC test was sponsored after its impact tests but if you look at the Thar (for which ESC was tested on two cars) it looks like the ESC test cars were reused for the impact tests.

This is something I hadn't quite considered. Thanks for bringing it to my notice. It further solidifies your initial argument.

I'm quite surprised that Tata hasn't sponsored an ESC test for the Nexon yet, even after they've made the system standard across all trims.

I'm more surprised that they don't advertise the 5-star rating for the Nexon anymore. While GNCAP doesn't document all the guidelines governing its modus operandi, given that it follows well established guidelines and procedures put forward by other NCAPs (mainly Euro NCAP), I wonder why Tata didn't file for retaining the rating of the pre-facelift Nexon, assuming such a provision exists in GNCAP? Euro NCAP has already outlined such a provision:

4.2.2. A ‘facelift review’ will be conducted by Euro NCAP to establish whether or not the original star rating can be transferred. This will comprise, in addition to those items examined in an
annual review (see above):
• A review of the changes that have been made to the vehicle.
• A review of in-house test data, where appropriate.

4.2.3. Application should be made even for ‘facelifts’ which have no influence, or a very minor influence, on the safety rating but where the appearance of the car has been altered or where the car is to be marketed as ‘new’. It is Euro NCAP’s intention to keep consumers informed
of the applicability of the rating and any changes which distinguish the updated vehicle from the original should be reported.

4.2.4. It is the responsibility of the vehicle manufacturer to approach Euro NCAP with information concerning facelift changes. If no information is received from the manufacturer, Euro NCAP may assume that the facelifted vehicle does not meet the requirements of the original star rating. The original star rating may no longer be valid and the facelifted vehicle may be eligible for assessment.

[ron178]

Originally Posted by rpm

I'm more surprised that they don't advertise the 5-star rating for the Nexon anymore.

I'm highly uncertain about this, and I have a feeling this might have more to do with the time elapsed since the test than with the facelift of the Nexon. Latin NCAP has mentioned the maximum time period for which a manufacturer can advertise ratings, in the FAQ section on its website. It actually depends on the protocol under which it was tested. Since this would not apply to the Global NCAP, I'm not sure where to find corresponding rules for the #SaferCarsForIndia program. In that way, it's also interesting that Volkswagen no longer advertises the Polo's 4-star rating (they barely did anyway) and neither did Toyota during the Etios' final years.

Originally Posted by rpm

Skoda being an international player, already has experience with even more stringent norms, yet isn't able to compete with homegrown products from a segment below— brands with significantly lesser international presence.

This is what I have a problem with. Of course, all of this is really just speculation since nothing can be said till the car is tested. However, the VW Group produced a 4-star car in 2014. If they produce a 4-star car in 2021 under the same test protocol, that's hardly progress. Just because the competition scored 3 stars, that doesn't make 4 stars good by 2021 VW Group standards, not to mention the limited availability of the full suite of side impact protection. That said, I appreciate the standard ESC and 3-point safety belts (the latter being a requirement for the 5th Child Occupant Protection star - I'm just speculating though). Also, I would consider 4 stars a huge plus point in segments where 5-star cars don't exist (the Renault Triber, Tata Tiago and Mahindra Thar are examples).

Originally Posted by rpm

...only tested rival being barely 3-star.

Here's something to ponder: the rival you mentioned scored 8.03 including the full SBR point (head, neck and chest were all orange, yellow or green thus qualifying for the mark-up for SBR). That would mean that the ODB test score is 7.03 out of 16 which should effectively place it in the 2-star category, unless the SBR points were not included in the evaluation for some reason (no mention in the PDF). I wonder why this is.

This was a very interesting conversation, and I'd like to thank you for the insightful information you provided, however I believe we have already far exceeded the maximum we can go off-topic on this thread. We'll continue this discussion on a new thread or avail of the direct-message feature, when the next round of #SaferCarsForIndia results come out. If rumour has it right, they will include the Renault Kiger (and hopefully more).

Thank you once again and stay safe.

[Mortis]

There is another possibility that hasn't been considered. Both the 4 star and 5 star versions of the Nexon were crash tested but they didn't bother to film and edit the 2nd test to save cash or because it was a visually minor difference or short time gap between tests or whatever and just ended up reusing the old video with slight color changes and edits.

[Lowflyer23]

Originally Posted by rpm

Wakey wakey, it's time to revive a dead thread!...
think of this speculation.

Not a Tata fanboy, far from that in fact. But I'd still give the benefit of doubt to the GNCAP considering it's a non profit organisation which picks up cars randomly from the assembly line for crash testing.

This code thingy could have been a fault of the chaps handling the YouTube channel. I believe they used the same thumbnail for both the videos whereas the cars were different (i.e. with different codes), correct me if I'm wrong.

[rpm]

Originally Posted by Mortis

There is another possibility that hasn't been considered.

I'll try to address as many points as I can in your speculation.

Originally Posted by Mortis

Both the 4 star and 5 star versions of the Nexon were crash tested but they didn't bother to film and edit the 2nd test to save cash or because it was a visually minor difference

They will have to go through the bother of filming the crash video regardless, because it is necessary and integral to their procedure of bestowing a crash rating. If they could get away with dummy readings and post crash measurements alone, the criteria would not necessitate looking for additional evidence for certain parameters, as well as defining certain parameters where the slowed down frames captured from the high speed cameras are necessary for assessment. Let me explain with an example.

UNSTABLE CONTACT ON THE AIRBAG
If during the forward movement of the head, its centre of gravity moves further than the outside edge of the airbag, head contact is deemed to be unstable. The score is reduced by one point. If for any other reason head protection by the airbag is compromised, such as by detachment of the steering wheel from the column, or bottoming-out of the airbag by the dummy head, the modifier is also applied. In case the head contacts any interior part of the car, excluding the rebound phase, the modifier will also be applied e.g. the dashboard.

Note: Head bottoming-out is defined as follows: There is a definite rapid increase in the slope of one or more of the head acceleration traces, at a time when the dummy head is deep within the airbag. The acceleration spike associated with the bottoming out should last for more than 3ms.The acceleration spike associated with the bottoming out should generate a peak value more than 5 g above the likely level to have been reached if the spike had not occurred. This level will be established by smooth extrapolation of the curve between the start and end of the bottoming out spike

Now, to check if the centre of gravity of the head moves further than the outside edge of the airbag, GNCAP also needs visual evidence captured by the high speed cameras. Since the crash happens too quickly to observe with the naked eye, they need the visual data from the slowed down frames captured by the high speed cameras to assess such a parameter. Further, to assess the condition of the head bottoming out, they need to assess the data when the head is deep within the airbag, meaning they again need visual evidence from the slowed down frames to correlate the data captured by the dummy itself. You'll see a distinct milliseconds reading at the bottom right of the crash video supporting this. So, filming it is almost a necessity, while saving the money (don't know about the estimate for the edit job) it would take to edit it for a YouTube video is not something I can conclusively comment on.

Your speculation about saving money on editing it for a YouTube video may stand, but the part about not filming it altogether seems difficult to buy into. Besides, even if the difference in footage was minor, as you say, they could at least have put the picture of the new Nexon (with the new code) undergoing the ODB test in their crash result pdf, or on their site. I assume that wouldn't need too much money to edit and include in the pdf or on their site when they probably would have already captured slowed down frames for assessing the vehicle in the hypothetical case of a new ODB test being conducted.

Originally Posted by Mortis

or short time gap between tests or whatever and just ended up reusing the old video with slight color changes and edits.

The time gap between the first test and the second test, is 45 - 26 = 19 weeks, or roughly 133 days, assuming my and ron178's interpretation of the code holds true. If you find any flaws or examples for which our current understanding of the code doesn't hold true, please point them out. I'll be happy to take a look and reconsider a better interpretation of the code. You may also present your own interpretation in case the current one seems unsatisfactory or incomplete.

Originally Posted by Lowflyer23

Not a Tata fanboy, far from that in fact. But I'd still give the benefit of doubt to the GNCAP considering it's a non profit organisation which picks up cars randomly from the assembly line for crash testing.

This code thingy could have been a fault of the chaps handling the YouTube channel. I believe they used the same thumbnail for both the videos whereas the cars were different (i.e. with different codes), correct me if I'm wrong.

I'd like to point out that the thumbnails for the YouTube videos are indeed different for the 4-star and 5-star Nexon. Instead of the thumbnails, the problem is with the entire clip of the ODB test in the 5-star Nexon video being the exact same as the 4-star Nexon, including the code that appears on the car in both the videos. Contrast this to an entirely different code on the MDB test, where the vehicle arrived at the lab 19 weeks after the ODB vehicle (assuming my and ron178's interpretation of the code is correct). If you go to GNCAP's channel, you'll find at the bottom left of both the videos that the clips used for ODB test are titled OD2618TNE1 in both these cases. Further, the crash result pdf and the GNCAP website also shows an image of the Nexon with the code OD2618TNE1 being crashed into the offset deformable barrier.

In this context, if I were to believe a slightly modified version of your speculation that the YouTube guy responsible for the GNCAP channel messed up and used the exact same clip (instead of the thumbnail— your original assumption) for both the tests, I would then also have to believe that the one who made the crash result pdf also messed up and used the wrong image, and so did the guy who uploaded the result to their site.

I would again like to reiterate that I'm not suggesting that the 5-star Nexon is actually 4-star. The only thing I'm alluding to, is the possibility of no ODB test being conducted for the 5-star Nexon. ron178's speculation includes an extra layer of no chassis changes being made to the Nexon, citing the provision for removal of knee modifiers without a crash test. Even then, he considers the possibility of changes being made to the dashboard. Of course, only Tata and GNCAP would know whether the cars from VIN MAT627165JLP51255 onwards entail structural changes or dashboard changes or just the addition of passenger SBR. None of us is suggesting that the 5-star rating awarded to the pre-facelift Nexon is wrong/incorrect/biased. Thanks to ron178's post, which documents the provision of a sled test for removing knee modifiers, provided that no structural modifiers (read body-shell integrity unstable, footwell rupture) have been applied, and no frontal protection airbag deploys incorrectly. If the provision for such a correction exists in the protocol, it is not wrong for the manufacturer to avail it to their benefit, especially since it is subject to being accepted by GNCAP, who, in turn, will only accept it if they are reasonably satisfied with the manufacturer's demonstration.

With all that being said, I would also like to reiterate that this is a speculation, because there is no way to confirm anything without GNCAP's or Tata's acknowledgement. By that logic, it would also apply to any speculation of a new Nexon with chassis changes being sent to the crash test, despite there being no video or photo evidence of the same being done. I and ron178 have tried our best to include all possible evidence to give some credibility to our speculation. It would help if members disagreeing with the speculation point out flaws in our interpretation and/or the evidence presented to make our case, which roughly translates to the interpretation of the code, evidence of the same clip being used for 5-star Nexon's ODB test, and the provision for removal of knee modifiers without a crash test. In the end, agreeing or disagreeing with the speculation is entirely upto the reader.

[Mortis]

Good points @rpm. I was just speculating and trying to provide an alternate explanation. The easiest way to clear this up is to email them or post on their twitter feed

[ron178]

Originally Posted by rpm

The car must score minimum 14.00/16 points in the offset deformable barrier test itself. Do note that this score only includes the ODB test, not the SBR. The 1 point allotted to SBR is a separate requirement independent of the ODB score.

Originally Posted by ron178

Here's something to ponder: the rival you mentioned scored 8.03 including the full SBR point (head, neck and chest were all orange, yellow or green thus qualifying for the mark-up for SBR). That would mean that the ODB test score is 7.03 out of 16 which should effectively place it in the 2-star category, unless the SBR points were not included in the evaluation for some reason (no mention in the PDF). I wonder why this is.

I'm sorry to revive a dead thread yet again but I'd like to correct a statement I made, regarding the exclusion of SBR points from the evaluation.

SBR points are indeed included in the evaluation. The scores printed in the grading table in the protocol are including any SBR points scored. This explains why the Renault Triber, the airbag-equipped Tata Zest and the Kia Seltos scored 4 stars despite their ODB test scores being in the 3-star, 3-star and 2-star categories respectively. So the Seltos, if tested before seatbelt reminders were mandatory, would be a 2-star car. Scary.

This could also imply that, say, the Ford Aspire (10.49) could be rated 4 stars when equipped with the government-mandated seatbelt reminders (assuming they meet Global NCAP requirements). I'm not sure if there are any other cars just bubbling under the 11.00-point mark that were tested without seatbelt reminders before they were made mandatory.

I think what the table should read is (including SBR points):
14.00 – 17.00 points (including 1 point SBR, no less) + 4ch ABS + ECE95* 5 stars
11.00 – 13.99 points 4 stars
8.00 – 10.99 points 3 stars
5.00 – 7.99 points 2 stars
2.00 – 4.99 points 1 star
0.00 – 1.99 points 0 stars
I'm still a bit unsure whether 5 stars requires 14.00 points including the SBR points or not. That is, if a vehicle crossed 14.00 points including 1 SBR point, would it be rated 5 stars? I'm led to think our old assumption regarding this was correct (14.00 without SBR). The culprit? The South African Honda Amaze, yet again. It scored 14.08 points with driver SBR. Now the 4-star rating could also be due to the full SBR point not being scored, but I think that if scoring the extra 0.50 passenger SBR point would mean a 5-star rating (after an MDB test of course), then Honda would have gone ahead and made it standard in South Africa because it's a small cost to pay for the publicity of a 5-star rating. Whether they were unsure of the MDB test result is not for me to say, though.

[rpm]

Originally Posted by ron178

......SBR points are indeed included in the evaluation. The scores printed in the grading table in the protocol are including any SBR points scored. This explains why the Renault Triber, the airbag-equipped Tata Zest and the Kia Seltos scored 4 stars despite their ODB test scores being in the 3-star, 3-star and 2-star categories respectively. So the Seltos, if tested before seatbelt reminders were mandatory, would be a 2-star car. Scary.

I had already figured out that this was the case back when Kia Seltos results dropped. However, I refrained from jotting down another post since you had requested that we postpone this discussion until the next round of GNCAP results dropped, for the risk of going far too OT. I'm glad that you have decided to prepone this discussion.

Originally Posted by ron178

This could also imply that, say, the Ford Aspire (10.49) could be rated 4 stars.........

Now this is something I had definitely overlooked. I had figured that the Triber was 3-star if not for SBR, but the fact that Ford could get a 4-star rating just by adding SBR gives me a new perspective on this car. I was miffed that Ford was going the Maruti way because the Aspire had an unstable body-shell integrity which was a downgrade from its predecessor. This paints the Figo/Aspire in an entirely new light for me. Thanks for pointing this out.

Originally Posted by ron178

I think what the table should read is (including SBR points):
14.00 – 17.00 points (including 1 point SBR, no less) + 4ch ABS + ECE95* 5 stars
11.00 – 13.99 points 4 stars
8.00 – 10.99 points 3 stars
5.00 – 7.99 points 2 stars
2.00 – 4.99 points 1 star
0.00 – 1.99 points 0 stars

I would agree that GNCAP's pdf leaves a lot of ambiguity with its scoring criteria. Instead of saying that it uses the frontal impact score for its star rating, it should say that it uses the overall score, including SBR, and the point limits apply to every rating, except the 5-star, for which, special requirements must be met.

Originally Posted by ron178

I'm still a bit unsure whether 5 stars requires 14.00 points including the SBR points or not. That is, if a vehicle crossed 14.00 points including 1 SBR point, would it be rated 5 stars? I'm led to think our old assumption regarding this was correct (14.00 without SBR). The culprit? The South African Honda Amaze, yet again. It scored 14.08 points with driver SBR. Now the 4-star rating could also be due to the full SBR point not being scored, but I think that if scoring the extra 0.50 passenger SBR point would mean a 5-star rating (after an MDB test of course), then Honda would have gone ahead and made it standard in South Africa because it's a small cost to pay for the publicity of a 5-star rating. Whether they were unsure of the MDB test result is not for me to say, though.

Our old assumption is correct. We have some examples to back ourselves up. Let me present to you, the latin NCAP tested Creta from 2015. This Creta was tested back when Latin NCAP had a protocol similar to the current GNCAP protocol, and assessed adult occupant protection from a total of 17 points (16 ODB + 1 SBR).



Now, this Creta scored an impressive 15.57/17 which is quite a bit more than the bare minimum threshold of 15.00/17 (including SBR) we had established earlier. The reason it still didn't get a full 5 star in AOP is because
1. It did not have passenger SBR, as evident from the pdf.
2. No side impact test was conducted, as specified by the pdf.

This Creta far exceeds the 14.00/16 threshold for ODB test, since deducting 0.5 points (for driver SBR) from its overall of 15.57, still leaves it with an impressive 15.07 for the ODB test. It just needed passenger SBR and side impact compliance, yet Hyundai chose not to sponsor further testing and/or add passenger SBR. This is despite the fact that NCAPs intimate manufacturers (according to protocol) of the result before they are released to the public, to allow manufacturers a chance to improve the crash rating of their product.

From this, I conclude that 1 point SBR, side impact compliance, 4 channel ABS, and ODB ≥ 14.00/16 are all mandatory requirements for a 5-star rating. If the car doesn't comply with all of these, the 5-star rating will not be awarded under the current protocol.


Here are the relevant parts of the latin NCAP protocol from back then.



In case you're wondering how am I sure that this protocol pdf from 2013 is the correct one, it's because from 2016 onwards, they started assessing AOP from a total of 34 points.

[ron178]

Originally Posted by rpm

Now this is something I had definitely overlooked. I had figured that the Triber was 3-star if not for SBR, but the fact that Ford could get a 4-star rating just by adding SBR gives me a new perspective on this car. I was miffed that Ford was going the Maruti way because the Aspire had an unstable body-shell integrity which was a downgrade from its predecessor. This paints the Figo/Aspire in an entirely new light for me. Thanks for pointing this out.

They've actually already added SBR as standard equipment since it's a Government requirement now. I wonder if they could get their ratings bumped up by having a newer car with SBR tested. Would certainly help sales considering the segment is almost full of 2 star cars (including SBR points). However, make no mistake; the Aspire is no yesteryear Ford. Even with 4 stars, the Aspire would still remain clearly different from the European and Latin American Ka Sedan structurally. The Ka may have scored 0 stars under both, the 2016 and 2020 LatinNCAP protocol but its bodyshell has shown less rupture than the Indian car in both tests (the frontal test was actually never repeated as I mentioned in an earlier post). That said, it's incorrect to assume that the old Figo, even with optional airbags, was necessarily safer than the Aspire, as I've seen many people do on this forum. (But I do see their point since the Figo was structurally similar to an older European Fiesta) A stable bodyshell and airbags don't necessarily guarantee a high rating as we've seen with the Honda Mobilio.

Originally Posted by rpm

Our old assumption is correct. We have some examples to back ourselves up. Let me present to you, the latin NCAP tested Creta from 2015. This Creta was tested back when Latin NCAP had a protocol similar to the current GNCAP protocol, and assessed adult occupant protection from a total of 17 points (16 ODB + 1 SBR).

I completely forgot that I could check 2010-15 LatinNCAP results for more examples. However, your example still doesn't clear up conclusively whether the minimum score in the ODB test for 5 stars is 14.00 or 13.00. One piece of evidence we do have is the Nexon's knee modifier removal that we discussed a month ago. If 14/17 including SBR was the minimum for 5 stars then Tata could've simply added passenger SBR and reached 5 stars. This in my opinion validates your claim.

Anyway, I'm getting a bit confused about the scoring protocol. I'm personally convinced now that Global NCAP does not involve in unfair practice (which was the reason I started this analysis of the test protocol anyway). However I have realised that reading the ratings isn't as simple as the star rating itself. Some cars with the same star rating under the same protocol (almost completely based on frontal impact) aren't equally crashworthy. I still have a few questions, and by this point I'm sure I must seem crazy for bothering so much about something trivial.

2.1 Points Calculation
A sliding scale system of points scoring has been adopted for the biomechanical assessments. This involves two limits for each parameter, a more demanding limit (higher performance), beyond which a maximum score is obtained and also a less demanding limit (lower performance), below which no points are scored. For the adult rating, the maximum score for each body region is four points. Where a value falls between the two limits, the score is calculated by linear interpolation.
For all tests that are part of the adult occupant protection assessment, capping limits are maintained for criteria related to critical body regions: head, neck and chest. Exceeding a capping limit generally indicates unacceptable high risk at injury. In all cases, this leads to loss of all points related to the tests. Capping limits can be equal to or higher than the lower performance limit, depending on the test.

Despite the last line, the capping limits for the head, neck and chest correspond to the lower performance limits for Global NCAP tests.

3.1.1 Head
3.1.1.1 Drivers with Steering Wheel Airbags
...
If there is hard contact, the following limits are used:
Higher performance limit
HIC36 650 Resultant Acc. 3 msec exceedence 72g
Lower performance and capping limit
HIC36 1000* Resultant Acc. 3 msec exceedence 88g
3.1.1.2 Drivers with No Steering Wheel Airbag
...
Lower performance and capping limit
HIC36 1000
Resultant peak Acc. 120g
Resultant Acc. 3 msec exceedence 80g
3.1.2 Neck
Higher performance limit
... [excuse me for not pasting these values since they get pasted incorrectly and I have to type them manually, but they are irrelevant here]
Lower performance and capping limit
Shear 3.1kN @ 0msec, 1.5kN @ 25 - 35msec, 1.1kN @ 45msec*
Tension 3.3kN @ 0msec, 2.9kN @ 35msec, 1.1kN @ 60msec*
Extension 57Nm*

3.3 Scoring & Visualisation
The protection provided for adults for each body region are presented visually, using coloured segments within body outlines. The colour used is based on the points awarded for that body region (rounded to three decimal places), as follows:
Green 4.000 points
Yellow 2.670 - 3.999 points
Orange 1.330 - 2.669 points
Brown 0.001 - 1.329 points
Red 0.000 points

Now, take a look at the Volkswagen Polo, Ford Figo and Maruti-Suzuki Eeco without airbags. All have at least one among head, neck and chest red. Red corresponds to 0.000 points for the specific body region, which means that the biomechanical values were higher (worse) than the lower performance limits. However, for critical body regions like the head, neck and chest, the lower performance limits are also capping limits. If you look at the bold part of the quoted portion of section 2.1 of the protocol, this means that the score should not only be 0.000 for the corresponding red region but should also be 0.00 for the entire test. We've seen this with more recent zero-star cars like the S-Presso (red passenger neck). However, the Volkswagen Polo without airbags, the first-generation Ford Figo without airbags and the Maruti Eeco have non-0.00 scores despite having a red head, neck and/or chest. Sure, the final score is zero stars which is probably the application of capping limits.

Now look at the Datsun redi-Go and the Renault Kwid (IV). Both have red chests. Their scores are actually in the 3-star region but it has been reduced to 1-star, presumably under this provision in Section 3.3 Scoring and Visualisation, under the aforementioned point-star table. Now according to the provision for capping limits for the chest (identical to the limit for a red chest) doesn't this mean that both these cars should have scored a neat 0.00 for the test since a critical body region passes its capping limit?

In order to avoid the highly undesirable situation of a vehicle gaining a rating of multiple stars when an important body region is poorly protected, the rating will be limited to no more than 1 star regardless of the total number of points scored.

...
For Adult occupant protection, the overall rating is based on the driver data, unless part of the passenger fared less well. It is stated that the judgement relates primarily to the driver. The adjusted rating for the different body regions is presented in a visual format of coloured segments within a human body outline for the driver and passenger.
...
For frontal impact, the body regions are grouped together, with the score for the grouped body region being that of the worst performing region or limb. Results are shown separately for driver and passenger. The grouped regions are:
- Head and Neck,
- Chest,
- Knee, Femur, Pelvis (i.e. left and right femur and knee slider)
- Leg and Foot (i.e. left and right lower leg and foot and ankle).

Now, let's try getting a qualitative score for the S-Presso
Head and Neck: Worst Performing Region: Passenger Neck: 0.000
Chest: Worst Performing Region: Driver Chest: 0.000
Knee and Femur: Worst Performing Region: Driver (one of them - I don't have the actual test values): non-zero
Leg and Foot: Driver right tibia
Modifiers:
Unstable Structure applied on 'Chest': Chest score remains 0.000
Unstable Footwell Applied on 'Leg and Foot': score reduces further
Variable Contact/Concentrated Loading on knees: 'Knee, femur and pelvis' score reduces further.
Capping limits exceeded: Yes, driver chest and passenger neck. Test score 0.00

Let's do the same for, say, a Datsun redi-Go:
Now, let's try getting a qualitative score for the S-Presso
Head and Neck: Worst Performing Region: Driver Head: non-zero
Chest: Worst Performing Region: Driver Chest: 0.000
Knee and Femur: Worst Performing Region: Driver (one of them - I don't have the actual test values): non-zero
Leg and Foot: Has to be driver since this isn't currently assessed for the passenger.
Modifiers:
Unstable Structure applied on 'Chest': Chest score remains 0.000
Unstable Footwell Applied on 'Leg and Foot': score reduces further
Variable Contact/Concentrated Loading on knees: 'Knee, femur and pelvis' score reduces further.
Capping limits exceeded: Yes, driver chest. Test score 0.00?
Assuming that capping limits are not valid, the Go's score would be restricted to one star (which it is) due to a body region scoring red, whether or not it is a critical body region or not. In that scenario, the Datsun redi-Go has its fair share of non-zero body regions that can take its score to 8.36.

What I want to ask is, have I misunderstood the meaning of capping limits? They seem to play no role at all. The S-Presso's score might be 0.00 after modifiers regardless of capping limits because other regions were poorly protected and might have been reduced to 0.000 after modifiers. But the VW Polo without airbags that I quoted earlier has scored better than the S-Presso for regions other than the head+neck and chest (where they are identical). In fact, it scores 5.42 points. Using the Datsun Go's logic, shouldn't the score be one star? It lies in the 2-star region but due to red body parts (here, critical, but assuming that it doesn't matter whether the red part is critical or not like in the Datsun Go) it should be limited to one star. Why, then is it 0 stars? Am I missing some crucial provision of the protocol?

An argument one might make is that cars without airbags score 0 regardless of injury measures. A quick look at older LatinNCAP results on the same protocol like this one disproves this. In fact, the tested Palio too had a red head but its score was in the 1-star region and that's what it got.

I'll take a look at some older LatinNCAP results sometime to check out different cases of scores around 14.00 to answer the question you posed. I'm fairly convinced that your conclusion is correct though. Thanks again for your time.

[rpm]

I'll be addressing the ODB score dilemma in this post, because I'll have to do quite a bit more reading to wrap my head around capping limits and scoring criteria.

Originally Posted by ron178

.......... However, make no mistake; the Aspire is no yesteryear Ford. Even with 4 stars, the Aspire would still remain clearly different from the European and Latin American Ka Sedan structurally. The Ka may have scored 0 stars under both, the 2016 and 2020 LatinNCAP protocol but its bodyshell has shown less rupture than the Indian car in both tests (the frontal test was actually never repeated as I mentioned in an earlier post). That said, it's incorrect to assume that the old Figo, even with optional airbags, was necessarily safer than the Aspire, as I've seen many people do on this forum. (But I do see their point since the Figo was structurally similar to an older European Fiesta) A stable bodyshell and airbags don't necessarily guarantee a high rating as we've seen with the Honda Mobilio.

That is true. I was miffed because Ford being the international player it is, could definitely have avoided the 1-point modifier to the chest region with their extensive know-how, and scored a 4-star even without SBR, yet they chose to compromise on that front. Not only that, they didn't even offer seatbelt pretensioners in a car that was the segment first in offering 6-airbags. It seems like a lot of wasted potential with that product to me.

Originally Posted by ron178

I completely forgot that I could check 2010-15 LatinNCAP results for more examples. However, your example still doesn't clear up conclusively whether the minimum score in the ODB test for 5 stars is 14.00 or 13.00. One piece of evidence we do have is the Nexon's knee modifier removal that we discussed a month ago. If 14/17 including SBR was the minimum for 5 stars then Tata could've simply added passenger SBR and reached 5 stars. This in my opinion validates your claim.

Even though I imparted the idea that we have more examples in the form of Latin NCAP results, I myself didn't do the due diligence before coming to the conclusion in the previous post. I did try to find more results, but the Latin NCAP site seemed a bit too cumbersome at first, so I just made merry with one result and came to a conclusion. You have rightly pointed out that it is not conclusive, and here, I'll post more examples which prove that my initial conclusion is incorrect. I apologize for rushing to a conclusion.

So, here we have some results in the 14.xx/17 category, and surprisingly, only one of them is a 4-star (because SBR didn't meet Latin NCAP requirements, and ECE95 wasn't conducted). Rest of them are 5-star despite scoring 14.xx/17 points, because they had SBR which complied with Latin NCAP standards, and they passed ECE95 as well.











From this, we can reasonably conclude that the 14.00 point threshold includes SBR score. A car with 13.00/16 ODB score can still get a 5-star rating, provided it scores a perfect 1 point for SBR, and passes ECE95 test, in addition to having 4 channel ABS. I would once again like to apologise for rushing to a conclusion.

Originally Posted by ron178

.......... I still have a few questions, and by this point I'm sure I must seem crazy for bothering so much about something trivial.

Not at all. You are asking those questions which none of us here had even contemplated. You helped solve the code and Nexon's mystery too. I think your contribution to this forum is invaluable. Please keep asking. If not me, some other BHPian better versed in this topic could offer you an interesting discussion, and possibly, a resolution. I too, will try and add to the discussion, if I'm able to figure out the questions you've asked. Thanks to you, we have (hopefully) resolved the ODB score dilemma too.

[ron178]

Originally Posted by rpm

So, here we have some results in the 14.xx/17 category, and surprisingly, only one of them is a 4-star (because SBR didn't meet Latin NCAP requirements, and ECE95 wasn't conducted). Rest of them are 5-star despite scoring 14.xx/17 points, because they had SBR which complied with Latin NCAP standards, and they passed ECE95 as well.

Thanks a lot, I meant to look up results later in the day but now I don't have to. This actually means a lot.

One, even if the remote possibility that two different Nexons were tested is true, the first one still becomes a 5-star car if it passes UNECE Regulation 95 and is equipped with a passenger-side seatbelt reminder. I wonder why Tata bothered having the knee modifiers removed then. the Nexon would have scored 5 stars (though by a very small margin) even with the modifiers applied. I'm not complaining, though - there's nothing wrong with having the modifiers removed but it's an added expenditure but not much gain since the masses normally don't look beyond the star rating - something Euro NCAP and ANCAP are being criticised for of late.

Secondly, that means a few more cars than we thought have crossed the minimum ODB score for 5 stars. The Toyota Etios Liva and the South African Honda Amaze come to mind. However, they have unproven side impact protection and poor side impact equipment so I won't jump to any conclusions. I'm already quite mad at Honda for providing a poor level of safety equipment (no ESC and not even thorax airbags as an option unlike the City which has them standard-fit) in an otherwise probably structurally sound car (I hold no responsibility for my statement if the Indian Amaze scores worse than the South African car). I think I now get why Global NCAP said that the South African Toyota Avanza 'just managed' 4 stars (11.57 including 1 point SBR) while the South African Honda Amaze 'scored a solid four stars' (14.08 including 0.5 point SBR) for adult occupant protection. Also, please don't take my fixating on the Amaze's test as fanboyism or anything of the sort, it's just that the brand lies in unknown territory despite being among the first to make airbags and ABS standard (a move they reversed with the Brio). The Mobilio didn't do too well but it was a cost-cut car anyway, and the 2008 City and Jazz that had standard airbags were never tested, so I'm very curious.

Another noteworthy point is that while the Indian Toyota Etios Liva scored 13.00 points (no SBR) (just 0.06 less than the Nexon), reaching the bare minimum ODB score for the 5th star, the South African car of identical specification scored 11.58 (no SBR) which is notably less and would not qualify for the 5th star even if other requirements were met.

Originally Posted by rpm

the Latin NCAP site seemed a bit too cumbersome at first,

I agree, but what I like about LatinNCAP's (and ASEAN NCAP's - though the rest of it is terrible) website is that the stars are sorted by colour for different protocol. I hope Global NCAP does this in India next year with the new protocol since I'm sure there'll be a few cars with very poor ratings and there will be people comparing the stars with older ones directly.

Originally Posted by rpm

Attachment 2184390

That kind of puts an end to "I'm sure it would have scored more stars with more airbags". However, I'm joking, I do encourage everyone to purchase cars with side impact protection.

Originally Posted by rpm

I was miffed because Ford being the international player it is, could definitely have avoided the 1-point modifier to the chest region with their extensive know-how, and scored a 4-star even without SBR, yet they chose to compromise on that front.

They really should have. The Aspire's 3-star result already ruined Ford's reputation in the eyes of many. What's more, I hear that the Aspire actually felt lighter built than the older Figo which had a stable structure, and that further reinforced people's gut feeling about the 'thunk of the doors' being related to crash structures. I had a hard time convincing myself that this was not the case - the final proof was when Mat Watson complained about the extremely tinny-sounding and feeling doors of the 2020 European Toyota Yaris - and Euro NCAP announced very soon after that that the Yaris was the first car to score 5 stars under the new, tougher protocol.

Originally Posted by rpm

If not me, some other BHPian better versed in this topic could offer you an interesting discussion

Honestly, I don't think anyone on the forum so far will have seen Global NCAP results with the depth that you have, at least not that I know of. I was elated when I saw your first post confirming that I wasn't the only one who had noticed the discrepancy in the Nexon's scores. Thanks once again for your patience and help.

[rpm]

Originally Posted by ron178

.............What I want to ask is, have I misunderstood the meaning of capping limits? They seem to play no role at all. The S-Presso's score might be 0.00 after modifiers regardless of capping limits because other regions were poorly protected and might have been reduced to 0.000 after modifiers. But the VW Polo without airbags that I quoted earlier has scored better than the S-Presso for regions other than the head+neck and chest (where they are identical). In fact, it scores 5.42 points. Using the Datsun Go's logic, shouldn't the score be one star? It lies in the 2-star region but due to red body parts (here, critical, but assuming that it doesn't matter whether the red part is critical or not like in the Datsun Go) it should be limited to one star. Why, then is it 0 stars? Am I missing some crucial provision of the protocol?

An argument one might make is that cars without airbags score 0 regardless of injury measures. A quick look at older LatinNCAP results on the same protocol like this one disproves this. In fact, the tested Palio too had a red head but its score was in the 1-star region and that's what it got.

The Polo's case is quite peculiar, now that you mention it. I don't have any clues about it. Unless this is another one of their mistakes, I don't have any explanation for this.

Originally Posted by ron178

...... I wonder why Tata bothered having the knee modifiers removed then. the Nexon would have scored 5 stars (though by a very small margin) even with the modifiers applied. I'm not complaining, though - there's nothing wrong with having the modifiers removed but it's an added expenditure but not much gain since the masses normally don't look beyond the star rating - something Euro NCAP and ANCAP are being criticised for of late.

I think Tata was just as confused about the scoring criteria as we were.
I agree that there is a lot more to a crash test than just the star rating. I'll highlight one such example (Skoda Kushaq Review) I've talked about before.

Originally Posted by ron178

Also, please don't take my fixating on the Amaze's test as fanboyism or anything of the sort, it's just that the brand lies in unknown territory despite being among the first to make airbags and ABS standard (a move they reversed with the Brio). The Mobilio didn't do too well but it was a cost-cut car anyway, and the 2008 City and Jazz that had standard airbags were never tested, so I'm very ccurious

I won't. Just like you are curious about the Amaze, I'm very curious about the 2.0 MJD and Tata Harrier speculation. We all have our cars/brands of interest. I'm also very curious about the Hyundai Verna (because my family bought one less than a year ago). I'm hoping that all the UHSS (according to RSR's body weld diagrams) makes for a 4-star rating at least.

Originally Posted by ron178

Another noteworthy point is that while the Indian Toyota Etios Liva scored 13.00 points (no SBR) (just 0.06 less than the Nexon), reaching the bare minimum ODB score for the 5th star, the South African car of identical specification scored 11.58 (no SBR) which is notably less and would not qualify for the 5th star even if other requirements were met.

Another noteworthy observation: the LHD Etios tested by Latin NCAP scored 12.86/17, and the pdf makes no mention of SBR.

Originally Posted by ron178

........ I hope Global NCAP does this in India next year with the new protocol since I'm sure there'll be a few cars with very poor ratings and there will be people comparing the stars with older ones directly.

This is bound to happen, no matter what. The common folk would directly compare star ratings of yesteryear with the current ones.

Originally Posted by ron178

That kind of puts an end to "I'm sure it would have scored more stars with more airbags". However, I'm joking, I do encourage everyone to purchase cars with side impact protection.

Does this 'higher airbag/variant would score better' train of thought trace its origin back to Autocar (Renuka and Hormazd) defending Kia Seltos' crash rating? I mean, it could theoretically work if side impact and side pole test were conducted and graded on a point based system which eventually contributed to the overall score, but as far as the current GNCAP protocol is concerned, I doubt anything but a knee airbag (as standard equipment) could sway the offset score in any significant way. Side airbags don't count for a test which just grades ECE95 on a pass/fail basis, which you can pass without airbags too (provided the structure itself is good enough to take the beating). This is one of the reasons I'm kind of mad at Tata for not providing side airbags even as an option on Tiago, Altroz, Nexon. Sure, the Altroz and Nexon passed ECE95 without them, but at least provide them as an option.

Originally Posted by ron178

....... What's more, I hear that the Aspire actually felt lighter built than the older Figo which had a stable structure, and that further reinforced people's gut feeling about the 'thunk of the doors' being related to crash structures. I had a hard time convincing myself that this was not the case........

Fortunately, I did not have any preconceived notions about 'thunk of doors' being related to the crash safety of a car (probably because I wasn't as keen an observer to notice such differences ). I had my apprehensions about about lower kerb weight cars, but that was already put to rest by the Etios long ago. I do, sometimes, check the kerb weights of international variants and compare them to the Indian variants. I don't recommend relying on them though. (Another bonus observation: The African Amaze differs in kerb weight from the Indian Amaze by a paltry 4-5 Kg)

Now, coming to capping limits and whether they are relevant at all.



The key to understanding this (according to me), lies in the fact that dummy data may be modified.

I have a hypothesis about how capping limits work, but I can't be certain about it. So anyways, it appears that there are two ways to arrive at a red coloured body region.

1. By exceeding the capping limit, as you have correctly deduced in your post.

2. When the modifiers applicable to that particular region, match or exceed the points scored for that region.

Let's take a look at some examples.

First off, we have the Jeep Wrangler.



I would like to establish how the scoring for full width frontal test works before we proceed. Bear in mind that this is not something I was able to gauge from the protocol, and neither did the method used for scoring in the ODB test work (worse performing region, be it the driver or passenger, is used for assessment). It could still be wrong, but I have found it to be pretty consistent when calculating the full width frontal test scores.

So, what we do here is, we sum the points for each region, for both the driver and the rear passenger, irrespective of who fared worse. Deductions are not subtracted from the score (my hypothesis is that they are already deducted from the relevant body region to arrive at the score for that region. Refer to point 2 for arriving at a red body region I have outlined earlier). We then add the scores for the driver and rear passenger, to get a total out of 32 points, and then divide it by 4 to arrive at the full width frontal test score outlined in the pdf. Alternatively, you could sum each body region for the driver and passenger separately, divide them by 2, then add them to get a total out of 16 points, and then, divide them again by 2 to get a total out of 8 points seen in the pdf.

With that out of the way, now let's look at the Jeep Wrangler. You'll observe that the chest of the rear passenger is colored red for the full width frontal test. In the deductions section, you'll see -2 points for shoulder seat belt load.

Now this modifier belongs to the chest region, according to ANCAP's protocol

4.2.2 Chest

Steering Wheel Contact (Driver)
Where there is obvious direct loading of the chest from the steering wheel, a one point penalty is applied.

Shoulder belt load (Driver and Rear Passenger)
Where the shoulder belt load filtered at CFC60 exceeds 6.00kN a two point penalty is applied.

My hypothesis is that, the Jeep Wrangler got some non-zero score for the chest region in the full width frontal test, but, because of modifiers, it got reduced to zero. In this case, since the actual score was non zero without the modifier, the penalty was only restricted to that particular body region, instead of all the body regions. In such a case, NCAP probably allows an overall non-zero score. Though, I'm not sure if it allows for multiple star rating in such a scenario.

Now, let's take a look at SsangYong Tivoli's full width frontal test.



Here, you will see that the head of the rear dummy is colored red. But what is different here, is that a -16 point modifier has been applied because the values exceeded the capping limits. Here, you'll see the loss of all points related to the test because a critical body region presented a very high risk of life threatening injury. However, full width frontal test seems to grade rear passenger and driver separately, as outlined before, so the points lost were only for the rear passenger score. You'll see that this holds true when arriving at the score for frontal full width test outlined in the pdf, using the method outlined before.

Let's take a look at the modifier applied in this particular test. This modifier is included under the Head region.

Exceeding forward excursion line (Rear Passenger)
The score is reduced for excessive forward excursion. Where the head of the Rear Passenger exceeds the 450mm or 550mm forward excursion line as defined in the full width test protocol, a 2 or 4 point modifier respectively is applied. The modifier can be removed when it is shown by means of numerical simulation or a sled test that the HIII-50M does not contact the front passenger seat when in the 50M seating position, or when the HIC15 value is below 700 in case of contact with the front passenger seat.

Now, there are two possibilities in this modifier. If the rear passenger exceeds the forward excursion line, a 2 or 4 point modifier will be applied. But for the SsangYong Tivoli, a 16 point modifier has been applied, which rules out this possibility. Besides, I could not find a capping limit for forward excursion line in their protocol. However, the second possibility presents a situation with a well established capping limit. HIC15 = 700 is the capping limit for the rear passenger , and since this capping limit belongs to a critical body region— the head, it allows for the possibility of stripping all the points for the rear passenger. You'll find this capping limit here.



I suspect that the head of the rear passenger dummy in the Tivoli exceeded the capping limit for the head region, meaning that it scored red without any modifiers. In such a case, NCAP may decide to strip all the points related to the test. But then, it still leaves why Tivoli got a 4-star rating up for debate. It does help explain (although, rather haphazardly) how and when capping limits work, but the issue of multiple star rating still stands.

There is another example of this multiple star rating issue that I found, but it isn't very conclusive (probably because I don't have enough data on it). The Mahindra Pik-up in Australia scored a red head, yet it got 2 stars back in 2011.



Now, the protocol back then was different, and not exactly the same as EURO NCAP protocol. I couldn't find much data on the protocol either, since it is very old. However, I did find some interesting bits, though I don't know which version of the ANCAP protocols they apply to. They are taken from the ANCAP site, from a pdf that is titled ANCAP notes on the assessment protocol.





With the star rating criteria outlined in the table, the 2-star rating of the Mahindra Pik-up makes sense, because it is above the points threshold for each respective category. Moreover, the last line in the last screenshot could theoretically explain why it could score multiple points despite a critical body region being colored red, although we do not know what the phrase 'some circumstances' in the last line entails. The HIC value given in Mahindra Pik-up's pdf exceeds 1000. I suspect that it may have got a red head without any modifiers (assuming it exceeded the capping limits), but I can't be reasonably sure because I do not have the protocol from back then.

I know this isn't a very satisfactory explanation for all your questions, but this is the best I could cobble together with my limited ability to interpret the protocol. It doesn't answer why Polo still got a zero star. Nor does it conclusively answer your doubts about interpreting capping limits. I do, however, hope that you find something useful in here that helps you on your quest to find answers, should you choose to pursue this topic. Also, I'll suggest that you look at ANCAP results too, if you wish to research further. Their pdfs are very detailed, even more so than EURO NCAP's. They offer a lot more data, as you have already witnessed with the Tivoli, Wrangler and Pik-up above.

Originally Posted by ron178

Honestly, I don't think anyone on the forum so far will have seen Global NCAP results with the depth that you have, at least not that I know of. I was elated when I saw your first post confirming that I wasn't the only one who had noticed the discrepancy in the Nexon's scores. Thanks once again for your patience and help.

I'd be lying if I say that I didn't feel outclassed by your posts. Your understanding of the protocol has been exemplary so far. It certainly exceeds my ability to comprehend the protocol. I wasn't expecting much of an audience when I wrote my initial post about the Nexon, much less a discussion which would actually solve the last number of the code, as well as what possibly happened in the Nexon's case. I was overjoyed to find not only an audience, but also assistance. Discussing with you has increased my knowledge about crash tests many fold, while also bringing out a lot more nuance than what my original post could ever hope to achieve. For that, I thank you. I thank you for the time and effort you put into crafting each and every post of yours.

[ron178]

Originally Posted by rpm

Does this 'higher airbag/variant would score better' train of thought trace its origin back to Autocar (Renuka and Hormazd) defending Kia Seltos' crash rating? I mean, it could theoretically work if side impact and side pole test were conducted and graded on a point based system which eventually contributed to the overall score, but as far as the current GNCAP protocol is concerned, I doubt anything but a knee airbag (as standard equipment) could sway the offset score in any significant way. Side airbags don't count for a test which just grades ECE95 on a pass/fail basis, which you can pass without airbags too (provided the structure itself is good enough to take the beating). This is one of the reasons I'm kind of mad at Tata for not providing side airbags even as an option on Tiago, Altroz, Nexon. Sure, the Altroz and Nexon passed ECE95 without them, but at least provide them as an option.

Actually, I was referring more to numerous people in the YouTube comments section, however it could be interpreted in your way too. As I said, I was joking. Side airbags are very important and in no way do I mean that they are useless if the frontal impact score is bad (as I've seen people say often). However, people who said that the Seltos would have got more stars with side airbags under the current protocol are mistaken. As far as Tata is concerned, it's important to note that ECE95 is a fairly basic test. It's more of a minimum regulatory requirement for markets than an actual NCAP test. Tata seems more concerned about the actual star rating. As long as the car scores 5 stars, you can't avoid people saying that 'the side structure is so strong that you don't need side airbags'. It's an unfair comparison, but it's still important to note that it would be extremely hard to pass something like an IIHS side impact (with the taller MDB) or even the small overlap test without side and curtain airbags, even then, the curtain airbags must extend far enough forward to prevent the dummy hitting the dashboard. Take a look at this test of a Volkswagen Beetle without curtain airbags (torso airbags didn't deploy) scoring poorly in a frontal test despite an acceptable structure.

Originally Posted by rpm

In this case, since the actual score was non zero without the modifier, the penalty was only restricted to that particular body region, instead of all the body regions. In such a case, NCAP probably allows an overall non-zero score. Though, I'm not sure if it allows for multiple star rating in such a scenario.

The problem is, we can't use other NCAPs here because they also conduct other tests. If a capping limit is exceeded, the score for that test is reduced to zero, but there are still other tests that can maintain a reasonable adult occupant protection score, which is what I think happened with the Tivoli. With Global NCAP, however, the score for the frontal offset test being reduced to zero implies the score for the full test is reduced to zero. Moreover, I'm not sure if the capping limits are the same as lower performance limits for other NCAPs. Global NCAP technical director Alejandro Furas said in an interview (please allow me some time for a citation) that for Global NCAP, 'a red body region corresponds to a fail in the UN95 test' (I'm not sure of the basis of the claim because of different test speeds). For developed markets, however, it is possible that the capping limits are higher than the lower performance limits, that is, the car could score zero for a body region (without modifiers) but still be above the capping limit and hence not have its score for that test reduced to zero. However, I will have to look at other NCAPs' protocol sometime to confirm this, as well as go over your post in detail. Edit: I read a portion of your post, and this does not seem to be the case at least as far as the Euro NCAP frontal full width test is concerned.

Originally Posted by rpm

I have a hypothesis about how capping limits work, but I can't be certain about it. So anyways, it appears that there are two ways to arrive at a red coloured body region.

1. By exceeding the capping limit, as you have correctly deduced in your post.

2. When the modifiers applicable to that particular region, match or exceed the points scored for that region.

It will take me some time (maybe even days) to read the rest of your post in detail, but I think you have answered my question.

In case of that Datsun redi-GO and the fourth Renault Kwid, it's possible that the chest scores were zero only after application of the unstable structure modifier to the chest. Hence, since dummy readings did not actually cross the capping limit, the test score was not reduced to zero. In the case of the Polo, however, it's possible that they simply printed the actual score but gave it zero stars because it crossed capping limits. Sure, it's inconsistent with later results, but it's still fair to the star rating. This means that all zero star cars including the S-Presso had actual biomechanical values (without modifiers) that crossed the capping limits. The Polo and first-gen Figo didn't even have an unstable structure modifier which makes it clear that the biomechanical values themselves crossed capping limits and the score should indeed be zero. Global NCAP reiterated on multiple occasions that it was the S-Presso's neck score that brought it down to zero stars, despite the chest also being coloured red. This could mean that the actual chest compression readings did not cross the capping limits, the chest was red only after the unstable structure modifier. Unfortunately, this also means that the S-Presso would likely still be a one-star car after adding a passenger airbag. Maybe pretensioners could help further, though.

To sum up, I think the zero-star cap is based on biomechanical capping limits, and the one-star cap is if a region crosses the lower performance limit after application of modifiers after application of modifiers.

However, I now have a new question, I think I should read the rest of your post in detail before asking it, but I'll ask it anyway.

In order to avoid the highly undesirable situation of a vehicle gaining a rating of multiple stars when an important body region is poorly protected, the rating will be limited to no more than 1 star regardless of the total number of points scored.
This assessment will be applied on the basis of dummy response alone, for any body region where there is an unacceptably high risk of life-threatening injury i.e. if the dummy response has exceeded the lower performance limit. The body regions which could give rise to a ‘star cap’ are the head, neck and chest.

This is inconsistent with my above observations, and I'm now baffled. I hope reading the rest of your post clears this up. The only possible explanation is that 'dummy response' includes modifiers, which is unlikely.

Edit: ANCAP's decision to award SUVs back then a default 16/16 for side impact does not sound fair.

Originally Posted by rpm

Now this modifier belongs to the chest region, according to ANCAP's protocol
My hypothesis is that, the Jeep Wrangler got some non-zero score for the chest region in the full width frontal test, but, because of modifiers, it got reduced to zero. In this case, since the actual score was non zero without the modifier, the penalty was only restricted to that particular body region, instead of all the body regions. In such a case, NCAP probably allows an overall non-zero score. Though, I'm not sure if it allows for multiple star rating in such a scenario.

I accept your statement that the score for the individual body region but not for the full width test gets reduced to zero if the region is coloured red only after modifiers. However, Euro NCAP (and ANCAP who has been using their protocol and sharing test results since 2018 - even from LHD cars) also conducts other tests, and hence the 'rating being limited to one star' would not apply to their protocol, unlike Global NCAP whose ratings largely depend on a single test. Rightly so, the adult occupant score is 50%. What caused the 1-star rating appears to be the poor safety assist score.

Originally Posted by rpm

So, what we do here is, we sum the points for each region, for both the driver and the rear passenger, irrespective of who fared worse. Deductions are not subtracted from the score (my hypothesis is that they are already deducted from the relevant body region to arrive at the score for that region. Refer to point 2 for arriving at a red body region I have outlined earlier). We then add the scores for the driver and rear passenger, to get a total out of 32 points, and then divide it by 4 to arrive at the full width frontal test score outlined in the pdf. Alternatively, you could sum each body region for the driver and passenger separately, divide them by 2, then add them to get a total out of 16 points, and then, divide them again by 2 to get a total out of 8 points seen in the pdf.

This seems correct for the frontal full width test. Based on your examples, it looks like a red head without modifiers means loss of all points for the test, and a red head only after modifiers means the score for that body region is zero. Since Global NCAP uses only the frontal offset test, a red head after modifiers limits the score to one star. I won't have much clarity on this until I inspect other NCAPs' protocols, which I don't intend to do, at least not anytime soon.

Thanks a lot for the data, it really did help.

[rpm]

Originally Posted by Nexusios1521

Could it be that just the added safety features and the side impact test were enough to improve the score and score an extra star? Global NCAP does not mention of Tata having made any structural improvements to the Nexon, thus requiring no retest(?).

Sorry for bothering you again, but I felt that it is necessary to update you with correct information now that we have a better understanding of the protocol. As it turns out, my understanding of the protocol back when I tried answering your query was incorrect. I would like to apologise for the same.

Originally Posted by rpm

To be eligible for a 5-star rating, certain requirements must be met. Let's have a look.

1. The car must score minimum 14.00/16 points in the offset deformable barrier test itself.............

So, for a 5-star rating, the minimum overall score needed by a car is 14 points ODB + 1 point SBR = 15.00/17 points.

This post is incorrect as of now, and the information presented here is also incorrect.

Originally Posted by rpm

........... You have rightly pointed out that it is not conclusive, and here, I'll post more examples which prove that my initial conclusion is incorrect. I apologize for rushing to a conclusion.

So, here we have some results in the 14.xx/17 category, and surprisingly, only one of them is a 4-star (because SBR didn't meet Latin NCAP requirements, and ECE95 wasn't conducted). Rest of them are 5-star despite scoring 14.xx/17 points, because they had SBR which complied with Latin NCAP standards, and they passed ECE95 as well.

Attachment 2184386

Attachment 2184387

Attachment 2184388

Attachment 2184389

Attachment 2184390

From this, we can reasonably conclude that the 14.00 point threshold includes SBR score. A car with 13.00/16 ODB score can still get a 5-star rating, provided it scores a perfect 1 point for SBR, and passes ECE95 test, in addition to having 4 channel ABS. I would once again like to apologise for rushing to a conclusion...........

This is the correct information.

So, to answer your query— yes, it is possible for the 4-star Nexon to become a 5-star car with just the addition of passenger SBR, provided that it passes the side impact test. Its overall score would be (13.56 + 0.5 passenger SBR) 14.06/17 in this case, but it would still be a 5-star car nonetheless. ron178's post aptly covers the most probable scenario of the current 5-star Nexon, that is, it had its knee modifiers removed following a sled test which increased its score from 14.06/17 to 16.06/17, but, even if Tata hadn't done that, and just added passenger SBR in addition to passing the side impact test, the Nexon would still get a 5-star rating, albeit with a slightly lower score of 14.06/17.

I would again like to apologise for answering your query with incorrect information stemming from my incomplete understanding of the protocol back then.