[Mi10]

Hi Gurus and fellow members,

I am back with one more list of bizarre questions ;-)

Consider the following scenario;

I have a Midbass that has sensitivity of 91 db (2.83V/1M)
This is rated at 160W RMS @ 4 ohm

and then i have a tweeter that has sensitivity of 93 db (2.83V/1M)
This is rated at 200W @ 6 ohms (maximum transient power handling of 1000W)

Now if i am going fully active (NO pasive crossovers)

Can someone tell me what is the power rating of the Amp is i should use here without Damaging the Tweeter!!


Thanks in advance

[DerAlte]

1. What are you trying to achieve - an SPL system?
2. With possible loss of auditory faculties (maybe even cerebral ones) or without?

[Mi10]

Originally Posted by DerAlte

1. What are you trying to achieve - an SPL system?
2. With possible loss of auditory faculties (maybe even cerebral ones) or without?

1. SQ System
2. Without losing organs

[abhibh]

Tweeters actually dont need a lot of power to play. So even if you have a 50x4 W RMS amp it can keep both midbass and tweeter happy.

Originally Posted by abhibh

Tweeters actually dont need a lot of power to play. So even if you have a 50x4 W RMS amp it can keep both midbass and tweeter happy.

thats a generalization thats dangerous to make - because you could have a small tweeter with those specs which is playing only above 4000hz whereby your statement could be true and this would be matched by a midbass/mid speaker which plays from where the sub stops up to around that 4000hz but to get higher end speakers that manage this is difficult

on the other hand you could have a large format tweeter which plays much lower ie from around 1500hz or so which would benefit from as much power as you can give it because it will be playing a lot of the midrange frequencies so a majority of the vocals will be emphasized there and to create a richer sound as opposed to having very thing vocals which will be prevalent if you dont power it properly you need to remember that a manufacturer is not mad to recommend the amount of power that they do for any particular speaker

for example you cannot run 50 watts into a dynaudio esotar II tweeter or even a morel supremo tweeter for example - those will require a bit more power than that because they are huge and definitely do play lower in frequency than your conventional small format tweeters

so for a tweeter of the caliber like i mention i certainly wouldnt use a 50 watt amplifier because you wouldnt even be making those start to work and you would definitely get disappointed in investing huge amounts of money into the speaker and having an amplifier thats underpowering it by a huge margin - and underpowering on its own wont damage a speaker but you will then try to pick up volume levels and possibly amp levels till clipping and distortion set in and hence damage the speakers whereas if they were adequately powered and you used restraint with your volume level and set the gain to never clip then the speaker will perform optimally at the volume level required

which is why i say you cannot make a generalization without knowing what speaker you are working with - to be honest i would use what the manufacturer recommends for a speaker - as i have said they make those recommendations based on what they have made that speaker to do

just my opinion

one other factor could affect your calculations ie where you intend to place the drivers - for example if the tweeters are mounted on the A-pillars they could become overbearing and you would need to still turn them down a bit whereas if they are mounted in the kickpanels then it makes more sense to match the sensitivities because its more likely that the midbass speaker would also be in the kickpanel or maybe in a door

if its in an a-pillar then match it so that the tweeter is possibly around 3db lower in output than the midbass because of the distances between them BECAUSE the sensitivity is calculated at a finite distance ie generally at 1 metre

BTW the biggest amp you could use without damaging the speaker is the manufacturers recommendation as long as you dont set the gains incorrectly and dont play the volume levels too high ie until you hear it distorting

the fact that the power is there doesnt mean yo have to be playing the system to its limits, i like to have loads of headroom and even if i play the system softly most of the time i would like to have loads of power on tap for the occasions when i do want to let my hair down and thats the time that underpowering any speaker will start to cause problems

[Mi10]

Thanks for that wonderful reply Nitn. That really cleared few doubts in my mind.

Now as stated in my example specs:
If i need to Match (or atleast come close to) the Sensitivities of the drivers,
Will my below assumption correct (theoretically atleast) ?

Midbass - 91 db (2.83V/1M); 160 rms @ 4 ohms
Tweeter - 93 db (2.83V/1M);200 rms @ 6 ohms

So to reduce 3 db i ll need to half the power
hence if i feed 100 rms @ 6 ohms to the tweeter its sensitivity reduces to 89 db.

And if mounted on the A-Pillar, On-Axis (considering the Midbass in the Kick panel) wont this sensitivities match (to some extent) ?

Also 100 rms @ 6 ohms translates into ?? rms @ 4 ohms

Thanks all for your support in assisting me in my naive queries

[navin]

Originally Posted by naughty001

for example you cannot run 50 watts into a dynaudio esotar II tweeter or even a morel supremo tweeter for example

the fact that the power is there doesnt mean yo have to be playing the system to its limits, i like to have loads of headroom and even if i play the system softly most of the time i would like to have loads of power on tap for the occasions when i do want to let my hair down and thats the time that underpowering any speaker will start to cause problems

1. I have driven the SEAS Milennium, ScanSpeak 9900 and Morel Supreme (the one with the 130 mm faceplate and offset dome) using a simple 6550 triode (PP) amp. I think the amp puts out about 10-12 W rms in that triode mode. No Problem. XO was around 2kHz in fact we were tinkering with the active XO to test the linear excursion of these drivers (DO NOT DO THIS AT HOME). These drivers retail for about 200 Euros each but manufacturers get them for about 1/3rd that price so they can fool around with these drivers with less care.

2. While I do agree that underpowering a speaker is more likely to result in a damaged tweeter when one goes active we have to understand that
(a) the tweeter amplifer is less likely to be pushed to overload since it sees about 10% of the musical engery (the rest being directed to teh woofer/mid amps)
(b) Tweeters are more likely to die from excursion than thermal power overload so 5W rms at 500Hz can kill a tweeter while 20W rms at 5000Hz might not.

Originally Posted by Mi10

If i need to Match (or atleast come close to) the Sensitivities of the drivers,
Will my below assumption correct (theoretically atleast) ?
Midbass - 91 db (2.83V/1M); 160 rms @ 4 ohms
Tweeter - 93 db (2.83V/1M);200 rms @ 6 ohms

Also 100 rms @ 6 ohms translates into ?? rms @ 4 ohms

P = VI, V = RI. db = 20 log V2/V1 or 10 log P2/P1.

2,83V @ 4ohms = 2W / @ 6 ohms = 1.33W
91db @ 2W = 88db @ 1W
93db @ 1.33W = 92db @ 1W

your tweeter is hence 4db MORE sensitive than your woofer. Placing it closer to you will only make that more apparent. if you are using teh same amp to drive both, pad your tweeter to taste. If you are using different amps then use the gain controls to set levels to taste.

[DerAlte]

Originally Posted by Mi10

...
So to reduce 3 db i ll need to half the power
hence if i feed 100 rms @ 6 ohms to the tweeter its sensitivity reduces to 89 db.

And if mounted on the A-Pillar, On-Axis (considering the Midbass in the Kick panel) wont this sensitivities match (to some extent) ?

...

1. Incorrect. The sensitivity of a driver is an inherent property, independent of the wattage that you manage to push through. If you reduce the power that you are feeding, the OUTPUT reduces, not the sensitivity. Even at that specific power feed, the driver will sound louder than a driver with lesser sensitivity
2. Since the frequencies dealt with by the tweeter and midbass are substantially different, it is incorrect to compare loudness like that. Again, sensitivities of the individual drivers will remain constant, you are talking about comparing actual loudness

[Mi10]

Originally Posted by navin

2. While I do agree that underpowering a speaker is more likely to result in a damaged tweeter when one goes active we have to understand that
(a) the tweeter amplifer is less likely to be pushed to overload since it sees about 10% of the musical engery (the rest being directed to teh woofer/mid amps)
(b) Tweeters are more likely to die from excursion than thermal power overload so 5W rms at 500Hz can kill a tweeter while 20W rms at 5000Hz might not.

P = VI, V = RI. db = 20 log V2/V1 or 10 log P2/P1.

2,83V @ 4ohms = 2W / @ 6 ohms = 1.33W
91db @ 2W = 88db @ 1W
93db @ 1.33W = 92db @ 1W

your tweeter is hence 4db MORE sensitive than your woofer. Placing it closer to you will only make that more apparent. if you are using teh same amp to drive both, pad your tweeter to taste. If you are using different amps then use the gain controls to set levels to taste.

Wow!. If only i had loved electricals & electronics engineering, when i studied it.

Navinji, if i am planning to crossover at 1.8 KHZ, the (fo of the tweeter is 700Hz, so i asssume xovering at one octave above is safe). and considering the fact that its rated at 200 rms @ 6 ohms, is it safe to feed it 75w rms(with gain settings to zero) each ?

So, in this case wont an Off-Axis placement of tweeter have better blend (with the Midbass)? I mean i dont want the tweeter to shout on my face (dont know how to put it )

Originally Posted by DerAlte

1. Incorrect. The sensitivity of a driver is an inherent property, independent of the wattage that you manage to push through. If you reduce the power that you are feeding, the OUTPUT reduces, not the sensitivity. Even at that specific power feed, the driver will sound louder than a driver with lesser sensitivity
2. Since the frequencies dealt with by the tweeter and midbass are substantially different, it is incorrect to compare loudness like that. Again, sensitivities of the individual drivers will remain constant, you are talking about comparing actual loudness

1. Yes sir, i get your point now.
2. I am not comparing the Loudness here - and as always, drivers of higher frequencies would have more sensitivity than drivers of lower frequencies. Just want to know if mix and match of drivers are reaosnably mated.(in terms of sensitivity)
But i thought the key here is SPL. Even though the sensitivity of the driver doesn not change, wont the SPL be less if i give lesser power ?

no Mi10 sir - you are not understanding what the sensitivity spec is. It is the amount of SPL a driver would give you from a specific distance away given exactly one watt of power. The given distance is usually one meter away

navins method of conversion is one hundred percent correct - i use a different methodology - mine is more of a shortcut (possibly you could call it a thumbsuck) but i would get similar results to him - but he is far cleverer than i am at this type of stuff ie the maths equations etc so i will go by his calculations as being spot-on

The sensitivity of a driver is an inherent property, independent of the wattage that you manage to push through

Deralte-ji - the sensitivity is actually dependent (as opposed to independent) on a specific amount of power ie one watt at one meter away

however if what you are trying to point out though is that it wont change according to power increases then thats one hundred percent correct

If you reduce the power that you are feeding, the OUTPUT reduces, not the sensitivity

100% correct sir - the sensitivity doesnt change according to power adjustments it is the output LEVELS

Since the frequencies dealt with by the tweeter and midbass are substantially different, it is incorrect to compare loudness like that.

with the utmost of respect (no sarcasm intended), not at all sir, since IMHO you want to match the amount of treble you output to be relative to the amount of midrange and midbass you output ... in fact its an aim of SQ to get a linear if not perfectly flat RTA graph so you would have more-or-less equal amounts of treble, midrange and bass - so how loud one speaker driver is has to determine how loud the others would need to be - though you have to take distance into account as well so that you are accounting for the arrival times of each frequency but as a start trying to use the sensitivities of the drivers to allow you to calculate what drivers to use as well how much power to give them is a start - you can then adjust levels for the distances thereafter but you have a basis to start from to plan the actual speakers and amplifiers you need to use to get a successfully matching system

i reiterate that this is just my opinion - anyone is free to disagree

if you are using teh same amp to drive both, pad your tweeter to taste. If you are using different amps then use the gain controls to set levels to taste.

since he has mentioned fully active i rather suspect that he will be using two amplifiers here ..... and hence it makes sense in this instance if the mid amplifier is stronger than the tweeter amplifier but the problem is that he seems to be using speakers that are not matched to each other ie the tweeter seems to be a higher end tweeter hence it is probably designed to be used with a stronger mid than he is planning to use anyway, It doesnt matter what he is specifically going to use but the specs tell us he should be using a weaker tweeter but again i rather suspect that the weaker tweeter will not provide a similar frequency response hence the mid needs to be capable of playing a very wide frequency range ie it should have a frequency response from around 50 to 60hz all the way up to around 8000hz or so +/-3db so he can bandpass it from whatever point he choses lower down to around 4000hz or so and then allow the tweeter to take over from there

again the idea is to chose speaker drivers that kind of will allow you sensitivities that allow their output levels to be similar given specific amounts of power you wish to use

it becomes easier if you did not have the amplifiers already and had to chose everything from scratch you could probably have a better matching system

(with gain settings to zero)

gain settings wont ever be zero anyway - because a gain is not a volume level - its a potentiometer that allows you to match the amplifiers input voltage to the headunits RCA output voltage. Gain settings would generally be between 250mv and 6v though these could change depending on the amplifier - i know some that can handle an input voltage of 9v and you would normally need a line driver for one of those. By reducing the gain you are kind of using it in similar fashion to a volume control by reducing the input voltage hence reducing the amount of power the amplifier will nominally produce

So, in this case wont an Off-Axis placement of tweeter have better blend (with the Midbass)? I mean i dont want the tweeter to shout on my face

that could help in terms of tonality but i dont know if it will help your soundstage a lot because a tweeter is generally very directional and this could adversely affect your soundstage and imaging or it might just possibly work - only way to find out is to experiment

So to reduce 3 db i ll need to half the power
hence if i feed 100 rms @ 6 ohms to the tweeter its sensitivity reduces to 89 db.

not at all - you still are not understanding sensitivity the sensitivity wont change - the eventual output level will change - that 93 db for the tweeter is if you feed it 1 watt of power and thats measured at one meter away - so if you fed it 100 watts of power then you wont decrease the sensitivity but instead you decrease the final output level - heres a tabulated version of how the power increases will affect the speaker

for the sake of the example lets assume a 4ohm speaker therefore the 2.83volts implies that this SPL number is attained at 2 watts

2 watts = 93db @1metre ( so we assume all the following are at one metre away
4 watts = 96db
8 watts = 99db
16watts = 102db
32watts = 105db
64watts = 108db
128watts= 111db
256watts= 113db

so feeding the tweeter approximately 100watts will make it give you around 110db in the frequency range it plays at so lets assume that the tweeter is generally flat to 20khz and you crossed over at 1500hz - so 1500hz to 20000hz is going to be more or less linear and all frequencies should be playing at around 110dbs when you play some pink noise at 0dbFS reference level

feeding the tweeter the full 200watts will be giving you close to around 112dbs of SPL but 75 watts will be giving you between 109 and 110db's

do a similar calculation for your midrange to see how much power can make that midrange speaker output around 109 to 110dbs - if thats within its rated power handling you are relatively safe to go ..... but with the provision that you are underpowering the tweeter so if you want it to play louder you shouldnt overdo it with the gain control to try and increase the output because you will cause the amp to clip since it cannot give you more than 75 watts of unclipped power (on paper .... in real world terms it CAN POSSIBLY give more but we wont account for that here it will just complicate matters)

given the above information i think its fair to assume that the tweeter you wish to use is mostly overkill ie its a case of trying to kill a cockroach with a shotgun and a lesser tweeter ie one that needs more like the 75 watts will do perfectly well instead so if you can find one like that which gives you the frequency response you need you are probably better off changing your plans on the tweeter to use even though you can safely use the one you are planning to use

also given that these figures are for a distance of one metre away you will need to adjust the tweeter levels down if its closer to you and maybe the mid will need more power to give the amount of output you desire for that particular frequency range if its further away than a metre - but its easy enough to subjectively judge this for yourself then adjust and experimenting with levels and maybe that off axis/on axis trick might be a lot of fun for you in terms of the installation

the idea with the calculations is to see if the drivers and amp you select can do the job you will be requiring them to initially do but they must be capable of adjusting up or down slightly to account for any fine tuning you need to do

again my entire post is just my opinion and anyone is free to disagree with me but thats how i do things and its served me well enough so far in getting close to audiophile levels of SQ in my car setups generally but i have been lucky that ive generally bought component speaker sets that are already matched and all ive had to do is leave out the passive crossovers and select amplifiers that work with the raw drivers so its been easier for me

Even though the sensitivity of the driver doesn not change, wont the SPL be less if i give lesser power ?

that statement is true enough and shows more of an understanding regarding the situation than the other statement regarding the sensitivity reducing - all you need to remember is that the SPL figure represents your output

[Mi10]

Ok here is the list for my Midbass:

160 Rms @ 4 Ohms
db Watt
91 2
94 4
97 8
100 16
103 32
106 64
109 128
112 256

Now Option 1: Tweeter X
200 rms @ 6 ohms
db Watt
93 1.33
96 2.66
99 5.32
102 10.64
105 21.28
108 42.56
111 85.12
112 170.24

Now Option 1: Tweeter Y
110 rms @ 4 ohms
db Watt
91 2
94 4
97 8
100 16
103 32
106 64
109 128
112 256


Seems like tweeter Y easier to match with the Midbass ?
and Tweeter X is surely an overkill


Assuming i take Tweeter Y, powering them at 75W RMS would get thme around 107 db. Now i need to power a minimum of 120W RMS so that my Midbass is around the 107 db mark. Am i right in my understanding now ?

[navin]

Originally Posted by Mi10

Navinji, if i am planning to crossover at 1.8 KHZ, the (fo of the tweeter is 700Hz, so i asssume xovering at one octave above is safe).

Yes but...
a. you need to crossover at 12db/octave or keep the power below 10Wrms.
b. When the manufacturer claims the tweeter can handle 200W they must also issue the condistions (slope and turnover point of crossover for example) under which these measurements were made.
c. It is not only the Fs of the tweeter but also the excursion capabilities of the tweeter (usually in the order of 0.2-0.5mm) that must be considered - thermal ratings are only a guide.

Also without FRD curves one does not know if there are any impedance changes, phase anamolies or breakup nodes that one need compensate for. So while a 1.8khz/12db XO will work it might NOT be absolutely correct and might not compensate for impedance, phase, amplitude anamolies.

There was a while paper many years ago but the guys at B&W on this but I cant remember where. Maybe one of you guys can google it - where is LBM when you need him?

[DerAlte]

Originally Posted by naughty001

... the sensitivity is actually dependent (as opposed to independent) on a specific amount of power ie one watt at one meter away ...

If 'Sensitivity' of the driver (which is a parameter measured @2.83V and 1 meter from the driver and declared) is x dB, this sensitivity figure (again, measured @2.83V and 1 meter) will not change under conditions that one is pushing (amplifier is pushing) varying power through the driver. 'Sensitivity' is a figure under standard measuring conditions, incl. a fixed frequency. If in reality there is a situation that the amp is producing 2.83V at the driver terminals at the same signal frequency, and you are measuring loudness at 1 meter, it should be the same as the 'Sensitivity' figure. At other voltage (power) levels / frequencies / distances, the measurement will produce a different loudness reading, but this will NOT be called 'Sensitivity'!

Originally Posted by naughty001

... you want to match the amount of treble you output to be relative to the amount of midrange and midbass you output ... anyone is free to disagree

Sir, it is not about agreement or disagreement - we seem to be on parallel tracks, and are getting lost in words! If one has to do something scientifically / systematically, the measurement, analysis and inference must have a rational, 'comparable' basis. If, say, one uses 1Khz as the test tone, do you think the mid-bass / mid-range / tweeter will sound equally loud? On the other hand, if one were to use a 20Hz to 20Khz sweep, what must one do to sense equal loudness at all frequencies, when drivers themselves have frequency dependent characteristics? Yes, the least we can do is to systematically establish a baseline for finding a matching pair from a (hopefully large) set of drivers.

But, if one doesn't have too many drivers to choose from, one has to use other methods to achieve matching - XO pad (passive systems) or amp gain (active systems). This method assumes the difference is within the limits of variation that the XO or amp can handle - which is usually the case.

[Mi10]

Originally Posted by navin

Yes but...
a. you need to crossover at 12db/octave or keep the power below 10Wrms.
b. When the manufacturer claims the tweeter can handle 200W they must also issue the condistions (slope and turnover point of crossover for example) under which these measurements were made.
c. It is not only the Fs of the tweeter but also the excursion capabilities of the tweeter (usually in the order of 0.2-0.5mm) that must be considered - thermal ratings are only a guide.

Also without FRD curves one does not know if there are any impedance changes, phase anamolies or breakup nodes that one need compensate for. So while a 1.8khz/12db XO will work it might NOT be absolutely correct and might not compensate for impedance, phase, amplitude anamolies.

1. Why not at 24db/octave ? wont that give us the headroom of giving more power?
2. & 3. Not sure sir. But what about Semi-Horn's ? Aren't they designed to take in more power and have better excursion capabilities than, say a conventional tweeter

Even if my Headunit has advanced adjustments to correct, cutoff Frequencies, Phase, Slope, etc..?

[navin]

Originally Posted by Mi10

db Watt
91 2

Tweeter X
db Watt
93 1.33

Tweeter Y
db Watt
91 2

Seems like tweeter Y easier to match with the Midbass ?
and Tweeter X is surely an overkill

Assuming i take Tweeter Y, powering them at 75W RMS would get thme around 107 db. Now i need to power a minimum of 120W RMS so that my Midbass is around the 107 db mark. Am i right in my understanding now ?

Arey baba if the Midbass is going to need 2 W to produce 91db it will need 70-75W to produce 107db. How did you get 120W?

Crossing the tweeter over with a steeper slope will improve power handling but will not double it (if you go to 24db/oct from 12db/oct). I assume you are refering to L-R XOs (for both 12 and 24db XOs) which offer a flatter power curve than Butterworth.

Horns and wave guides do increase sensitivity but also limit dispersion (which in some cases/applications is a good thing). However the do nothing for the power handling.