Preamp gain vs. poweramp sensitivity


Hello,

Just purchased an Art Audio Carissa which has input sens. of 600mV (very high). At first, I was thinking that I would want a high-gain pre because the Carissa puts out only 16W/ch and could benefit from some preamp gain.

However, I believe upon considering the sensitivity that a high-gain pre is NOT desired because:

1) I would need to open the VC only a bit before the Carissa was driven to full power.

2) For the above reason, the high-gain is wasted anyway because the pre wouldn't be able to put out what it's capable of without overdriving the Carissa.

Do I have this straight? The relation of gain to voltage is not clear to me. The preamp is specified as outputting 2V yet, with a standard 2V source, that is actually unity gain! So what exactly does "20db of gain" in a preamp mean?

Thanks for clarification on this question.

Paul
paulfolbrecht
Paul
Good question.You can look at my questions I just had with swapping out andf old Denon for a buddy with Maggie 3.6's whose input sensitivity must have ben really hot because now new higher powered Bryston (which meets THX therefore 2MV standars for input sensitivity) requires much more of the volume dial (which I told was in uitself not a bad thing in that pre's will wrork beter if they get turned up and don't just use 7'oclock to say 10 o'clock).But freind I had was freaked that he went from alteral bi-map 150 stereos (thuis a totasl of 225 watts totalk output) to a the Bryston which benches near 300 wpc.But he had to put his Mac 712 pre at 12 o'clock high to get what he got at 9 0'clock from the Denons.Bryston very helpful eplaining unity gin to me and I'd have to know what the Mac and Deono's were set at to understand it.But still it was just a difference that through him.Harmonics ,well everything sounded beter with Bryston and when turned up to 3'clock ones reeas began to bleed.But alwaysd the neuraoitic he says new CD's sopund great but his Blue Note CD's purchased in 80's sound like they were recorded in Rudy Van gelders parents living room (hold they were actaully recorded in his parents living room) and now sound like like crap.Tried to explan to hoim that iuncrease of quality from medum current mid fi to a true audiofile amp would no longer allow his amp to cover up the crap.Now his Magnepans SOUND like superior speakes whereas before I got depressed everytime we played music on his system and it sounded so bad.I thought "Why is everybody saying Magnepans are maybe the best bang for the buck in Hi end unless you needd massive volume or foundation shaking bass".Now I know.
But your question about the 20 db of gain is in intruiging to me as well.
G'luck
chazzbo
what exactly does "20db of gain" in a preamp mean
The pre amplifies the input ~100 times. Assuming this is voltage gain, of course.
As to the 2V in 2V out, remember the pre has a volume control. So it attenuates the output as required.
Check the rated output impedance of the pre & compare with the input of the amp.
Gain and sensitivity are actually 2 ways of talking about the same thing. One can be calculated from the other.

Sensitivity tells you how much input voltage it takes to drive the amp to full power, 600 mV is actually on the low side as many amps take 1.5 (1500 mV) volts or so to drive them to full power.

Gain tells you how much a device multiplies the input voltage. It is expressed either as a ratio of output to input (Vout divided by Vin) or using a log scale of dB.

gain in dB = 20 log (Vout/Vin)

20dB of voltage gain is actually a factor of 10, not 100.

A preamp with 20dB of voltage gain will take a .2 V input signal and multiply it times 10 to make it 2 V if the volume is all the way up.

Your amp takes 600 mV to produce 16 watts.

It takes about 11.3V out of an amp to produce 16W into 8 ohms. (power = voltage squared divided by ohms)

Your amp has a gain of 11.3 divided by .6 = 18.8 or expressed in dB = 20 log 18.8 = 25.5dB

The confusion with the 2V output ratings is because sources and preamps are rated differently. The 2V from a source is the maximum level it can produce. The 2V spec from the preamp was with a specified input level with the volume all the way up and not the maximum that it can produce, which will usually be at least 10V and usually more.

If the source did hit a peak of 2 V and the volume was all the way up with a 20dB (times 10) preamp then it would try to put out 20V. Since your amp only needs .6V for full volume it would clip. For this reason most line level preamps are actually used with their volume controls less than maximum and are actually attenuating (reducing) the level of the signal. It's not as bad as it seems since the average output level of the source is much, mush less than the maximum it is capable of.
Herman,

Thanks so much for that great explanation! Exactly what I was looking for.

I have to ask you now - would you suggest a passive linestage with this setup? My source's output imp. is 300ohm and the amp's is a very high 180K - this seems like a good case for a passive pre to me. Agree?

In other words, I am now concluding that ANY active pre, much less a high-gain one, makes little sense here. Impedance matching would be the only possible reason for an active preamp and that would not be an issue at all, certainly not with a passive pre with a typical 10K output impedance.

What do you think?

Thanks again.

Paul
Paul, FWIW, I don't believe your amp will see the output impedence of your source. It will see the output impedence of your (passive) line stage. Does the 'passive' line stage you are considering have a constant output impedence, or does the output impedence change with the position of the volume control. Some do, some don't and it can make a difference (but with and input impedence of 180K it might not). It could also affect you choice of IC's and the lengths you could use (short is always good with passives).

Lastly, I observed a lot of folks going to passives with a lot of initial satisfaction over what appears to be a gain in detail, clarity, etc, but concurrently I've seen quite a few change back (including me) feeling that they lost some dynamics with the passive. Try before you buy if you can.
Most passives do vary OI, you're right, with 10K being the top end. Into 180K, I think that'd still be fine.

I have tried a PLS on my system with the previous amps, which have only half the input impedance, and are less sensitive, and it seemed very good to me, including the dynamics.

Active preamps just seem like so much overkill when you've got a strong source and good impedance on both ends.
I agree with the responses about the passive. While your setup is certainly a great textbook match for a passive, some will still prefer an active stage. The good news is you can get a used passive and sell it at no loss if you don't like it. Personally, I've always like one with the right setup. There is usually a Placette or 2 for sale here and they retain their value very well. Give it a try.
May I borrow this thread? Perhaps Herman can clear my head. I have a stereo amp that is rated at 250 watts per channel in to 8 Ohms with 26.6 db of gain. What is the amps sensitivty?
On the other hand I have an amp with .8V sensitivity and puts out 225 watts (tested at closer to 300) onto 8 ohms. Can you calculate the gain. Thanks in advance.
Ok, Paul. Sorry for stealing your thread. Perhaps you might be able to confirm whether I'm correct in assuming that an amp with 26.6 db of gain has a sensitvity of 2.66 V for full out put?
I don't mind the bogarting of the thread. It's all good. I haven't run your #s but if you used Herman's equation it must be right.
250 watts per channel in to 8 Ohms with 26.6 db of gain.

26.6 dB..... 10 to the power of (26.6/20) = gain of 21.4

250W into 8 ohms takes 44.7V because ((44.7 squared)/8) = 250

44.7V divided by 21.4 = sensitivity of about 2.1V

.8V sensitivity and puts out 225 watts (tested at closer to 300) onto 8 ohms.

300 watts into ohms takes 49V

49V divided by .8V = gain of 61

20 log 61 = 35.7 dB
Clio, I think I did yours incorrectly. It may be capable of 300W, but if the rating is 225W with a .8V input, to get 300W would take an input voltage higher than .8V.

To get 225W into 8 ohms takes 42.4V

42.4 divided by .8 = a gain of 53

20 log 53 = 34.4 dB
Thanks for the update Herman. I suspected the gain was very high, just didn't think it was over 30db.

What I can say about gain is the improvement it can make to your system. I have an ML amp and 326S preamp, and I can set the preamp to o gain, 6db, 12db, and 18db. Each step up from zero improves the sound, I like the 12db, but I recently tried the 18Db, and what I have experienced is a more relaxed sound, more open, more dynamics in the bottom end, and just a more pleasant and natural sound without any downsides. 12Bd sounds more sharper and detailed but not as natural as the 18db, So gain does make a sonic impact which is why buying a new preamp with more gain than you had before will sound different for sure and most likely better. My vocals are now in the room with body and openness. I guess much less like a recording. 

@paulfolbrecht Yes, you were thinking about this right. Excellent numeric breakdowns by @herman!

Sensitivity in power amps is a voltage input number that has to be referenced against a specific output power level (225 Watts in your case) to resolve its voltage gain - usually it’s referenced against "full rated continuous power" (for a sinusoidal wave), regardless of what the amp actually measures at - the 300 Watts bench results are irrelevant to its sensitivity rating.

Sensitivity in speakers is a dB output number that’s usually referenced against an input of 2.83 Volts (one Watt into an 8 ohm load, or 2 Watts into 4 ohms).

Voltage multipliers have a constant factor of 20 when you calculate out their dB differential (dB = 20 * log(V1 / V2)), whereas power multipliers only have a factor of 10 (dB = 10 * log (P1 / P2)) - why is that? Because when you keep the resistive load constant (as in the downstream audio component), increasing the voltage also causes a proportional increase in current - and these increases both multiply together to create power. That’s why bridging a "high current" stereo amp (bridging always doubles its output voltage) can elicit a QUADRUPLING in power - as long as it can supply the extra current, exactly proportional to the doubled voltage :)

You amp is indeed unusually high in both gain and sensitivity. You would be wise to avoid preamps with high gain (anything above 14dB). However, I don’t think you necessarily need to stick to passives. When you combine high gain preamp w/ high sensitivity amp & high sensitivity speakers, you will likely face a high noise floor (white noise, hissss). I like Tannoys with lots of tube power, so I’ve been there before.

Tube amps tend to have a higher sensitivity (lower voltage required to hit full output) than solid state amps, though of course there are exceptions. I had an 1950s/60s Eico HF-87 tube amp that was rated at 35 Watts/ch and had a sensitivity of 0.38V. It had a deceivingly high gain (33dB). This is incidentally the same gain as my Rogue Apollo tube monoblocks, which are rated at 1 Volt for full 250 Watts output power. These amps have a noticeably higher gain than any other amps I’ve used! When you see a 12AX7 in V1, get ready for a lot of gain :) 

My passive LDR Tortuga Audio preamp is not lacking in any area. Try one and see for yourself.

So gain does make a sonic impact which is why buying a new preamp with more gain than you had before will sound different for sure and most likely better. 

That is simply incorrect. More gain from that component may have improved the sound in your system, or perhaps you are confusing louder with better, but there is absolutely no consistent correlation betweeen increased gain and better sound.