[Shiwik]

I tried searching through the forums for a solution and could not find one. For that reason i am posting a new one. Mods, please move this to the right place if needed and as appropriate.

I have installed a 14.5 inch roof mounted display(worldtech) in my car and it is connected to the car head unit(pioneer) which plays DVDs and USB. While the engine is off, the display is perfect. However when i switch on the engine and start accelerating the car, there are horizontal lines that are rendered on the over-head display. The display on the car head unit is perfect, so i suspect there is something else is happening. We have tried the following -

1. Change the video cable to another one
2. Provided grounding for the power supply to the roof mounted display unit
3. Additional grounding at the head unit

Unfortunately none of them worked and both the vendor (the shop where i got this unit fixed) and I are clueless on what needs to be done here. Anyone faced a similar issue or could think of what could be causing this, i am desperately looking for help here.

Advance thanks

[GTO]

Your solution might lie here:

Originally Posted by decode

Hi Kamal,
There is nothing much to worry here. This is a common problem with pioneer AVH.

It is simple,Go to your camera setting , change the video input mode from Auto to your camera system (NTSC, PAL etc..) and it should work fine. All the best

[joybhowmik]

Originally Posted by Shiwik

could think of what could be causing this, i am desperately looking for help here.

Originally Posted by GTO

Your solution might lie here:

Try the video out setting which GTO suggests, hope that's a quick fix.

From ^^^ description of your problem , it looks like this appears when you are running the car, not when the device is on battery.

It's possible this is being caused by a ripple in the input voltage. Taking the output from a DC/DC type SMPS (if you find one ) can help you.

Like this unit
DLS-1212DC_Datasheet.pdf

You can contact this company here

Or other companies like it.

[Shiwik]

Originally Posted by GTO

Your solution might lie here:

I tried all options mentioned in this thread to no avail. However this has educated me on this issue reasonably well. I am fairly confident now that the fluctuation in voltage causes these horizontal lines. Not sure what the next steps are as i am technically challenged.

Originally Posted by joybhowmik

Try the video out setting which GTO suggests, hope that's a quick fix.

From ^^^ description of your problem , it looks like this appears when you are running the car, not when the device is on battery.

It's possible this is being caused by a ripple in the input voltage. Taking the output from a DC/DC type SMPS (if you find one ) can help you.

Yeah it behaves this way only after the engine is on. Like I mentioned above, it does seem like the voltage fluctuation after the engine is started. I spoke to the vendor from whom i got it fixed and he has asked me to get it back to him to get this fixed. However, if it indeed is a fluctuation issue, i am not sure what help he could provide.

Anything else i could do? I scanned amazon.in and ebay.in for a converter but could not find any for 12V-12V. Any suggestions?

BIG THANKS to the two of you for getting me to understand the issue here. All this while I was wondering if i got a defective product.

[joybhowmik]

Originally Posted by Shiwik

Any suggestions?

The only one I have ....

Originally Posted by joybhowmik

Like this unit
Attachment 1289985

You can contact this company here

[DerAlte]

1. What car do you have? I assume you have a petrol engine

2. The thread title says "Roof-mounted Display showing vertical lines" but you have written "when i switch on the engine and start accelerating the car, there are horizontal lines that are rendered on the over-head display"!!! Which is the correct statement

3. Which HU (make, model) do you have in your car?

4. Vertical lines are an indication of LCD part failure. These lines will stay whether you provide a signal or not

5. Horizontal lines are an issue with noise in signal, which I suspect is alternator or ignition coil noise induced into the video signal

6. That means either your HU (video out) is misbehaving, or the video cable doesn't have sufficient shielding. The 2nd problem is solved easily by getting a good quality shielded cable, as opposed to chalu Video cables with yellow RCA plugs costing a couple of 100. Whether the HU is a culprit or not, only the manufacturer's service center will be able to say. But they have to keep the HU in your car to be able to observe the fault - won't appear if you take it out of your car

[Shiwik]

Originally Posted by DerAlte

1. What car do you have? I assume you have a petrol engine

I have diesel engine car.

Originally Posted by DerAlte

2. The thread title says "Roof-mounted Display showing vertical lines" but you have written "when i switch on the engine and start accelerating the car, there are horizontal lines that are rendered on the over-head display"!!! Which is the correct statement

My mistake about the title. It is indeed horizontal lines and appears only when the engine is on and keeps fluctuating based on the acceleration/deceleration.
There are no vertical lines (anyway to change the title?)

Originally Posted by DerAlte

3. Which HU (make, model) do you have in your car?

I have a pioneer touch screen (without bluetooth). Can't quite remember the model number. It used to run well in my earlier car (again diesel) that had a 9" carbon DVD overhead unit. There were no issues of horizontal lines then. The current one is 14.5" display only.

Originally Posted by DerAlte

4. Vertical lines are an indication of LCD part failure. These lines will stay whether you provide a signal or not

Excuse the confusion, there are no vertical lines. It is only horizontal wavy lines when the engine is on and fluctuates in thickness when the car is running

Originally Posted by DerAlte

5. Horizontal lines are an issue with noise in signal, which I suspect is alternator or ignition coil noise induced into the video signal

I feel the same too as the display works very well when the engine is off and the power is entirely from the battery.

Originally Posted by DerAlte

6. That means either your HU (video out) is misbehaving, or the video cable doesn't have sufficient shielding. The 2nd problem is solved easily by getting a good quality shielded cable, as opposed to chalu Video cables with yellow RCA plugs costing a couple of 100. Whether the HU is a culprit or not, only the manufacturer's service center will be able to say. But they have to keep the HU in your car to be able to observe the fault - won't appear if you take it out of your car

Like i said earlier, this used to work very well in my earlier car and even now it works well when the engine is off. We even tried alternating the display between AV1 and AV2 modes to no avail. I am assuming the cable to be about a couple of hundred rupees as it has the typical yellow plugs on either sides and i was not charged much for the installation. Is there any specific cable that you would recommend. Also if i have to ask the vendor to try a few things, is there anything that i have missed above that you would want me to ask him or do.

[DerAlte]

Originally Posted by Shiwik

... It is indeed horizontal lines and appears only when the engine is on and keeps fluctuating based on the acceleration/deceleration. ... Also if i have to ask the vendor to try a few things, is there anything that i have missed above that you would want me to ask him or do.

Must be alternator noise.

1. Get a better video cable, or get one made in SP Lane (SJP Road) with dense shield

2. Ask the installer to use a filter in the 12V line going to the monitor (this will be a ferrite bead with wire looped around the ferrite core). This is available as a component in SP Lane

[joybhowmik]

Originally Posted by joybhowmik

Like this unit

Originally Posted by DerAlte

Ask the installer to use a filter in the 12V line going to the monitor (this will be a ferrite bead with wire looped around the ferrite core).

DerAlte ,

Choke/ferrite bead is a cheap solution compared to a DC/DC SMPS.

My understanding is that the choke filters out the high frequency noise in the power signal.
In comparison a DC/DC SMPS can filter out a wider range of frequency/harmonics - but of course this depends on the individual unit design.
(do correct me if I am factually incorrect)

two noob questions :
a) Would the ferrite bead work better when engine rpm @ 3K vs when @ 1K?

b) If budget is not a constraint: for someone wanting crystal clear display when parked or when accelerating , would SMPS be the better option ?

[DerAlte]

Originally Posted by joybhowmik

... Choke/ferrite bead is a cheap solution compared to a DC/DC SMPS. ...

The best solution is always the one that doesn't bring in more problems. The ferrite bead is what has been used in millions of TVs and monitors. A DC-DC converter brings in it's own failure points.

Does it make sense to use a 12V to 12V DC-DC converter for the monitors which have built in SMPS? Even TVs and monitors have SMPS / DC-DC converters built-in, the way this monitor would have. Monitors need at least 3 voltages to be generated from 12V - 3.3V, 5V and a much higher voltage for the back-light. One doesn't use analog power supplies in consumer equipment nowadays - other than hi-fi amplifiers.

Originally Posted by joybhowmik

... My understanding is that the choke filters out the high frequency noise in the power signal.
In comparison a DC/DC SMPS can filter out a wider range of frequency/harmonics - but of course this depends on the individual unit design.
(do correct me if I am factually incorrect) ...

You are factually correct (other than "wider range of frequency/harmonics" - which is same as "high frequencies"), but ... perhaps you should consider that the noise he is suffering from is deadlier superimposed as induced EMI on the video signal, rather than the power supply.

Originally Posted by joybhowmik

... a) Would the ferrite bead work better when engine rpm @ 3K vs when @ 1K? ...

I will understand your question better if you could transpose the figures to EMI on the video signal. Since the noise is originating from the alternator (brushes), and the alternator is always higher RPM than the engine (being driven by a small pulley by the belt), it is significantly present from idling till redline. The ferrite bead is f agnostic, the DC-DC converter has f limitations due to its component values). Anyhow, the DC-DC converter part is moot here.

Originally Posted by joybhowmik

... b) If budget is not a constraint: for someone wanting crystal clear display when parked or when accelerating , would SMPS be the better option ?

No, it is not - it would be redundant.
* This is not ye olde analog TVs that would get it's major scan components affected by EMI: a scooter, autorickshaw or moped passing by could make you miss that crucial wicket or goal!
* The monitor already has SMPS inside
* It is the signal which is never too well protected from EMI
* The noise on the power line sometimes affects the logic circuits via induced EMI on power supply (for example, welding machine arc nearby), but these turn up as small flashes, not constant horizontal lines on display

[joybhowmik]

Originally Posted by DerAlte

I will understand your question better if you could transpose the figures to EMI on the video signal. Since the noise is originating from the alternator (brushes), and the alternator is always higher RPM than the engine (being driven by a small pulley by the belt), it is significantly present from idling till redline. The ferrite bead is f agnostic, the DC-DC converter has f limitations due to its component values). Anyhow, the DC-DC converter part is moot here.

Thanks Der Alte. Very lucidly explained. But I still have a niggling doubt about the kind of ferrite bead to be used. I ask a patient reading to this rather simplistic chain of thought...

Assuming a 3:1 ratio for crankshaft pulley to Alternator Pulley, then the AC current frequency that's generated at a 2 Pole alternator...
@ 1000 RPM : (1000*3) *2 / 120 = 50 Hz
@ 3000 RPM : 150 Hz.

This AC is of course rectified as it should be.
If I am not mistaken, the rectified current waveform will still have noise at the above frequency (and higher frequencies).
Further, its this noise which induces the EMI effect in the power line to the device in question. Correct me if I am wrong, but IIRC, the induced current will also be at the same frequency (not unlike current induced in transformer windings - per Faradays law).

So that means the ferrite bead ought to be most effective (high resistive losses) at precisely this frequency range.

Is there a specific type of ferrite bead which the OP should get based on his specific need?


TDK_InCompliance_Aug2010.pdf

All Ferrite Beads are not created equal
Understanding the Importance of Ferrite Bead Material Behavior

By Chris T. Burket, TDK Corporation
Published In Compliance Magazine, August 2010

another question: From above, and the ferrite beads seen attached to laptop adapters etc one would assume their operating range (resistive range) is 50 Hz to a 60 Hz - maybe 200 Hz. However as a review of available literature shows, the resistive range is measured in MHz or even GHz. So, why would someone even bother to use such a filter outside it's known resistive range?

[DerAlte]

Originally Posted by joybhowmik

... I still have a niggling doubt about the kind of ferrite bead to be used. ... So, why would someone even bother to use such a filter outside it's known resistive range?

It would be of great help if you relate to the problem correctly: EMI, not ripple in the rectified DC output as you are reckoning with. EMI is typically in the high 100s of KHz to mid 100s of MHz range. The source of the RF noise is micro-arcing at the brushes in the alternator.

EMI is common-mode noise in the shielded cable connecting Video-out of HU to Video-in of monitor. The audio equivalent of EMI in HUs and amps is "alternator whine" due to bad wiring practices and inferior RCA cables.

All good cables used in high-speed digital communications - USB, Firewire, monitor cables etc. - have a built in EMI filter; looks like a cylinder in line with the cable.

[joybhowmik]

Originally Posted by DerAlte

EMI is typically in the high 100s of KHz to mid 100s of MHz range.

Thanks! Understood and can relate this to the higher frequency noise which I observed in an oscilloscope when hooked to an alternator.

All good cables used in high-speed digital communications - USB, Firewire, monitor cables etc. - have a built in EMI filter; looks like a cylinder in line with the cable.

Yes.. in other words the ferrite bead... but is there a specific type or product or SKU based on specific need? Or, are all ferrite beads the same in operation?

Originally Posted by joybhowmik

Is there a specific type of ferrite bead which the OP should get based on his specific need?

[Shiwik]

The problem is finally solved. I did not use anything additional as suggested above. I took it to Tata Motors and the first thing they did was to check the voltage from the battery when the engine is off and then tried it after the engine in on. It displayed 13.9 and 14 V respectively. So fluctuating voltage was not being considered as an issue.

When they opened up the wiring, the first thing the technician asked me is "did you do the wiring". Apparently the wire for supplying the power that was used was a double wire (excuse me if i am not using the right terminologies). I meant, the two poles of the wire are fused together in one casing. Internally there were separate casing but were bundled into one large case.

What i was told was, the earthing wire generally needs to be a separate wire and needs to be as small as it can be. That was the only change he made. Disconnected the current earthing and created a new one which was significantly shorter. That is all it took.

What i realized out of this entire exercise is i am lot more literate on these concepts now and i definitely want to thank the two of you spending time in understanding my issue and suggesting work-arounds.