How to find the peak decibel gain of the highest gain channel of a high gain amp?

[thinkingchicken]

The reason I don’t understand your interest in exceeding this level of gain (115 or 140) is because in real world scenarios using a typical output guitar pickup, the signal level into the amplifier is already too high for this level of gain because clipping occurring and reducing gain.

How to find the peak decibel gain of high gain amplifiers? You don’t have an education level with substantial understanding of the concepts or mathematical experience needed to calculate.

That's why I asked people who have the experience. The genre that I'm trying to experiment with, which is harsh noise "music" , there is no such thing as pleasant sounding distortion like when you listened to Eric Clapton. I'm not after those kind of distortion, I'm after a more extreme version of guitar distortion. Hence, more gain and more dB gain but without these diode clipping.

[thinkingchicken]

The reason I don’t understand your interest in exceeding this level of gain (115 or 140) is because in real world scenarios using a typical output guitar pickup, the signal level into the amplifier is already too high for this level of gain because clipping occurring and reducing gain.

This is a very important point. At a certain level of amplification (gain) the signal will clip. This is not considered a bad thing as most guitar players like the way this sounds. So, as you add gain you keep pushing the guitar's signal into this upper wall. But there is also something else going on.

Riding along with a guitar's signal is another signal. It is background noise. It is produced by the electronics in the guitar itself. There is not much one can do to eliminate this noise short of cooling the guitar and cable to absolute zero. Normally it is much much weaker than the output of a guitar's pickups. So, it is not very noticeable. But then gain is added...

At low amounts of gain the guitar's signal may be say 100db louder than the background noise. So, it does not compete much with the guitar's signal. Now a bunch of gain is applied. Both the guitar's signal and the noise are amplified. And the guitar signal starts to clip. The guitar's signal is not getting any louder but the noise is. The range between the noise and the guitar signal is being compressed. So, now the guitar's signal may be only 12db louder than the background noise. You still don't hear the noise much as long as you keep playing. But when you stop playing the noise is very bothersome. Thus many high gain amps will include noise gates that mute the amplifier when the guitar signal stops.

If the gain is increased even more then the noise itself will start clipping. It is now just as loud as the guitar. And at this point no one can probably tell the two signals apart. It also probably does not matter what the guitar signal is because the noise will more or less drown it out. This is why you don't see guitar amps with say a 1000db of gain. It is pointless to increase the gain beyond a certain limit. My guess is that limit is very close to the gains reported by circuit simulation at the beginning of this thread.

Mike

If I have to take a guess, the limit of wave clipping is around 105-110 dB. And beyond that, the clipped waveforms will not change anymore. Is that correct? Please correct me if there is an error.

[teemuk]

Sigh. I feel I'm repeating myself.

There are various clipping thresholds. Some may be as low as 300mV (e.g. germanium diode), some as high as >150V (e.g. plate clipping of high voltage tube circuit). These are very much defined by the circuit.

There can be variance in amplitude of the input signals e.g. "line level", "guitar level", etc.

Gain is just "gain", IOW amplification factor/ratio. It's effects to clipping distortion are undefined unless we also specify input amplitude and clipping threshold.

[LOUDthud]

I have measured gain of a high gain guitar amp I built. It was 105dB IIRC to a 4 Ohm load. I used an Audio Precision System 1.

Setup:
Generator: Sweep Frequency, 80Hz to 10KHz. Output about 10uV IIRC. Output: single ended floating ground.
Measure: Voltage, DbG (Db in reference to Generator)
Set graph limits to capture graph.

The System takes an RMS measurement of the output on the measurement side, then calculates gain in reference to the output of the generator. Peak gain was in the 3KHz to 5KHz area. All controls all the way up. You have to watch as the generator sweeps so you don't get clipping on the amp's output. There was a change in noise when turning up the Gain control between 1st and 2nd stages telling me I could reduce noise by lowering the pot's resistance.

None of the people who have played this amp have even turned the Master Vol past 5 (except me). So maybe a better measure of gain would be just from the input to the top of the Master Volume control, but that wouldn't tell you what the clipping level is. I usually set the power amp gain to 100 (40dB). If I put a preamp with 100dB of gain in front of it, I'm pretty sure it would oscillate when the Master is on 10.

[R.G.]

So you're insisting that I'm an actually a person who have experiences in building amps and pedals. Cool, can't help with that but it's fine. It's up to you.

No, I'm telling you why you're not understanding the valid answers you've been given, and explaining why you might be getting answers that you don't like. As well as providing somewhat subtle pointers to how you could get to where you could understand a bit more.

As we have discussed previously, an open loop op amp have more gain than the maxed out highest gain channel of Marshall JVM410. So if someone tried to experiment with an open loop op amp by making it into the form of a solid-state amp, the end product will be just a noise generator? Correct?

It's not that simple. The end product is not easily predictable. Depending on the layout and shielding around it (remember that I referred to the construction details earlier?) it might oscillate at one frequency, which could be in the audio band, or ultrasonic, even RF, which would sound... silent. It could wander back and forth through several frequencies. It could be sensitive to the position of objects around it (remember theremins??) It could lock up so it passes nothing. It could be a mountain of hiss. It could produce square waves like a comparator. It could do some combination of all of those.

I urge you to try some of the circuits out. All of the videos you're referring to are the product of experimentation to get a certain setup to produce a sound that the maker - and you - happen to like. The pedals that produce controllable noise/feedback/oscillation are the produce of informed experimentation.

I'm still a little baffled about what it is you want us to tell you. Concepts? How to build one? How to choose boxes to make a certain sound? How to make odd-sounding experimental music setups? It's very unclear to me and to others if I'm reading their responses correctly. Can you state in a single sentence what your ultimate goal is?

[thinkingchicken]

Sigh. I feel I'm repeating myself.

There are various clipping thresholds. Some may be as low as 300mV (e.g. germanium diode), some as high as >150V (e.g. plate clipping of high voltage tube circuit). These are very much defined by the circuit.

There can be variance in amplitude of the input signals e.g. "line level", "guitar level", etc.

Gain is just "gain", IOW amplification factor/ratio. It's effects to clipping distortion are undefined unless we also specify input amplitude and clipping threshold.

Ok, bear with me for seconds. I re-read all of our convos, and lets see if I get it right already.

The more the amplification factor and the lesser the clipping thresholds, the faster the waveforms become distorted. Correct me if I'm wrong.

The tube amps like Marshall JVM410, EVH Stealth, MI Audio Megalith Beta, etc. all have more decibel gain than, say, a distortion pedal like Metal Zone, simply means the clipping thresholds of these high gain tube amps are so large that they take time to distort compare to a distortion pedal like the Metal Zone. IOW, distortion pedals distorted earlier because the diodes or transistors make the clipping thresholds much smaller although the voltage gain is small compare to the tube amps where it is just raw gain.

Let's say, if I still wanna own a custom solid-state amp with a peak decibel gain much higher than those tube amps. For example, at 180 decibels. At certain point, maybe if we turn the gain knob pass 5, this solid-state amp with a peak decibel gain at 180 decibels will be nothing more than a big noise generator. But it is no a big deal as I am experimenting with noises like those synth guys do. But isn't it if we're using a noise gate, the noises will be eliminated?

[thinkingchicken]

I have measured gain of a high gain guitar amp I built. It was 105dB IIRC to a 4 Ohm load. I used an Audio Precision System 1.

Setup:
Generator: Sweep Frequency, 80Hz to 10KHz. Output about 10uV IIRC. Output: single ended floating ground.
Measure: Voltage, DbG (Db in reference to Generator)
Set graph limits to capture graph.

The System takes an RMS measurement of the output on the measurement side, then calculates gain in reference to the output of the generator. Peak gain was in the 3KHz to 5KHz area. All controls all the way up. You have to watch as the generator sweeps so you don't get clipping on the amp's output. There was a change in noise when turning up the Gain control between 1st and 2nd stages telling me I could reduce noise by lowering the pot's resistance.

None of the people who have played this amp have even turned the Master Vol past 5 (except me). So maybe a better measure of gain would be just from the input to the top of the Master Volume control, but that wouldn't tell you what the clipping level is. I usually set the power amp gain to 100 (40dB). If I put a preamp with 100dB of gain in front of it, I'm pretty sure it would oscillate when the Master is on 10.

Interesting. The Crate BV300H amp that have eight 12AX7s tubes in the preamp, according to https://www.manualslib.com/manual/62232 ... v300h.html, the decibel gain (not sure if it is the peak decibel gain or not) is 105 dB too. The Dual Recto is also 105 dB according to the first link I have shared. Anyway, is the 105 dB figure come from the preamp alone or with the power amp?

Maybe the typical peak decibel gain of high gain tube amps are 105 decibel?

I hope I have the money to buy those similar setup that you used.

Did you turn the gain knob to max? If the gain knob and MV knob both at max, I believe the result will be chaotic. Typical guitarists tend to hate that, and synth guys sure love that noises.

[thinkingchicken]

So you're insisting that I'm an actually a person who have experiences in building amps and pedals. Cool, can't help with that but it's fine. It's up to you.

No, I'm telling you why you're not understanding the valid answers you've been given, and explaining why you might be getting answers that you don't like. As well as providing somewhat subtle pointers to how you could get to where you could understand a bit more.

As we have discussed previously, an open loop op amp have more gain than the maxed out highest gain channel of Marshall JVM410. So if someone tried to experiment with an open loop op amp by making it into the form of a solid-state amp, the end product will be just a noise generator? Correct?

It's not that simple. The end product is not easily predictable. Depending on the layout and shielding around it (remember that I referred to the construction details earlier?) it might oscillate at one frequency, which could be in the audio band, or ultrasonic, even RF, which would sound... silent. It could wander back and forth through several frequencies. It could be sensitive to the position of objects around it (remember theremins??) It could lock up so it passes nothing. It could be a mountain of hiss. It could produce square waves like a comparator. It could do some combination of all of those.

I urge you to try some of the circuits out. All of the videos you're referring to are the product of experimentation to get a certain setup to produce a sound that the maker - and you - happen to like. The pedals that produce controllable noise/feedback/oscillation are the produce of informed experimentation.

I'm still a little baffled about what it is you want us to tell you. Concepts? How to build one? How to choose boxes to make a certain sound? How to make odd-sounding experimental music setups? It's very unclear to me and to others if I'm reading their responses correctly. Can you state in a single sentence what your ultimate goal is?

Just wanna hear a comparison of distortion sounds between high gain tube amps for example those with peak decibel gain beyond 110 dB like JVM410, EVH, Megalith Beta, etc. and high gain solid-state amps, preferably with comparable peak decibel gain. IOW, preferably, solid-state amps that use no clipping diodes at all but just raw gain to achieve clipping.

The likely result will be, all of them will become noise generators. But if we use noise gates, the noises will be minimized. Right? So, as noises have been minimized, we can compare the distortion of a maxed out JVM410 and a maxed out solid-state amp. Generally, tube amps sound pleasant while solid-state amps and pedals tend to sound harsh. For example, compare the tube tone of Matthias Jabs from the Scorpions with the solid-state guitar tone of Dimebag.

Is there a difference between solid-state noise generators and tube noise generators like those from Metasonix? Perhaps, tube-based noise generators have less pronounced high frequencies compare to the solid-state noise generators.

[LOUDthud]

OK, I found my notebook that contains details on the amp and printouts of the graphs (and I didn't remember correctly). This goes back to December 2018. The peak gain of the amp with all controls on 10 is 110dB at 5KHz, 83dB at 50Hz, 103dB at 20KHz. With all controls on 5, gain is only 61dB (@5KHz). With all controls on 10 except Gain and Master on 5, gain is about 77dB (@5Kz). Digital files of the gain plots are across town at my shop.

The power amp gain is only 30dB to a 4 Ohm load. The amp clips just slightly at 100W so I call it 90W. I named it MC90.

If you can find an Audio Precision System 1, get the software and a computer with it. Software was written for Windows 98. There is an old version of the software that uses a PC parallel port. Consider it worthless.

[thinkingchicken]

OK, I found my notebook that contains details on the amp and printouts of the graphs (and I didn't remember correctly). This goes back to December 2018. The peak gain of the amp with all controls on 10 is 110dB at 5KHz, 83dB at 50Hz, 103dB at 20KHz. With all controls on 5, gain is only 61dB (@5KHz). With all controls on 10 except Gain and Master on 5, gain is about 77dB (@5Kz). Digital files of the gain plots are across town at my shop.

The power amp gain is only 30dB to a 4 Ohm load. The amp clips just slightly at 100W so I call it 90W. I named it MC90.

If you can find an Audio Precision System 1, get the software and a computer with it. Software was written for Windows 98. There is an old version of the software that uses a PC parallel port. Consider it worthless.

Really fascinating. So, the peak decibel gain is varied depending on the frequency response of the preamp and as long as the gain knob isn't turned all way up to 10 the decibel gain will be lower. Now I just knew that the preamp tend to have much more gain than the power amp too.

Have you measured other amps too?

I thought Audio Precision System is just a software where you can simply install it on your computer, but it is actually a set of a big kit. https://www.valuetronics.com/product/sy ... %20testing.

[R.G.]

Just wanna hear a comparison of distortion sounds between high gain tube amps for example those with peak decibel gain beyond 110 dB like JVM410, EVH, Megalith Beta, etc. and high gain solid-state amps, preferably with comparable peak decibel gain.

I think that if you had asked your original question that way, you might have received responses pointing to examples of distortion.

IOW, preferably, solid-state amps that use no clipping diodes at all but just raw gain to achieve clipping.

Why no diode clipping? Is there a particular reason other than what you've read on the internet?

Generally, tube amps sound pleasant while solid-state amps and pedals tend to sound harsh. For example, compare the tube tone of Matthias Jabs from the Scorpions with the solid-state guitar tone of Dimebag.

That is the perceived internet wisdom. Like most internet "wisdom", it is not necessarily so. There has been a great deal of work on making pedals sound not-harsh, and some solid state amps that are difficult to distinguish from tube amps by listening. Can you prove that statement by measurement, or is it only by what you heard on the internet?

Is there a difference between solid-state noise generators and tube noise generators like those from Metasonix? Perhaps, tube-based noise generators have less pronounced high frequencies compare to the solid-state noise generators.

It's difficult to make noise generators of the high-gain type be consistent from one unit to the next, let alone in a different technology. It's even harder to make accurate statements about the general sound of noise generators.

But let's think more philosophically. You're stating a bias against solid state. Can a solid state amp >ever< produce good guitar sound? I've had long discussions on this from people whose mantra was "you can't get tone from a stone". Are you in that camp? Can a tube amp's sound be recorded and sent through a solid state amp and still sound good (assuming it ever did in the first place)? What about a tube preamp through a solid state amplifier that (1) does or (2) does not clip?

[LOUDthud]

Really fascinating. So, the peak decibel gain is varied depending on the frequency response of the preamp and as long as the gain knob isn't turned all way up to 10 the decibel gain will be lower. Now I just knew that the preamp tend to have much more gain than the power amp too.

Have you measured other amps too?

I tried to measure an Express clone I built on a Traynor YBA-1 chassis. Couldn't get it to stop oscillating even after lowering the B+. The Express really has a sweet spot just below 400V B+.

Built a clone of a Rockaforte Custom 80 on a JCM 800 2203 chassis with the Marshall transformers. Probably some plots on the computer at the shop. IIRC gain was about 100 dB.

The amp before the MC90 was the MC1250. IIRC the gain was about 100dB in the 5F6 mode (no cold clipper) and only slightly higher in the 2204 mode. Originally that amp had a "Bedroom" mode with clipping diodes. Nobody liked them so I took them out.

The AP System can be tricked into measuring sag induced compression. Look at this thread: http://ampgarage.com/forum/viewtopic.ph ... 91#p249291

You setup a single frequency sweep where you sweep the generator level starting with a small signal. You can set the AP to automatically turn the generator on at the beginning of the sweep, or you can just leave the generator on and the system will adjust the level to whatever the sweep demands. If you set the amplitude of the generator so the amp clips hard and the power supplies sag before the sweep, the AP will measure the gain with a small signal before the power supply has a chance to recover when the sweep starts. Run it again but with the generator off before the sweep and you'll see the difference in gain that compression makes.

[thinkingchicken]

. *edited for double posting*

[thinkingchicken]

Really fascinating. So, the peak decibel gain is varied depending on the frequency response of the preamp and as long as the gain knob isn't turned all way up to 10 the decibel gain will be lower. Now I just knew that the preamp tend to have much more gain than the power amp too.

Have you measured other amps too?

I tried to measure an Express clone I built on a Traynor YBA-1 chassis. Couldn't get it to stop oscillating even after lowering the B+. The Express really has a sweet spot just below 400V B+.

Built a clone of a Rockaforte Custom 80 on a JCM 800 2203 chassis with the Marshall transformers. Probably some plots on the computer at the shop. IIRC gain was about 100 dB.

The amp before the MC90 was the MC1250. IIRC the gain was about 100dB in the 5F6 mode (no cold clipper) and only slightly higher in the 2204 mode. Originally that amp had a "Bedroom" mode with clipping diodes. Nobody liked them so I took them out.

The AP System can be tricked into measuring sag induced compression. Look at this thread: http://ampgarage.com/forum/viewtopic.ph ... 91#p249291

You setup a single frequency sweep where you sweep the generator level starting with a small signal. You can set the AP to automatically turn the generator on at the beginning of the sweep, or you can just leave the generator on and the system will adjust the level to whatever the sweep demands. If you set the amplitude of the generator so the amp clips hard and the power supplies sag before the sweep, the AP will measure the gain with a small signal before the power supply has a chance to recover when the sweep starts. Run it again but with the generator off before the sweep and you'll see the difference in gain that compression makes.

Thank you for explaining how the AP System can be used to measure gain. I'll try to look for one if these things are available in the SE Asia.

Never thought that Trainwreck Express has so much gain. Did the tremendous gain mainly come from the power tubes? Because I heard when it comes to these Trainwreck Express amps, it is the power amp that is the first to distort then the preamp. This B+ thing is the power supply, right? Seems like the more the power supply, the more the sound distorted when only raw gain is used without the clipping diodes. So, you just measured the power supply but not the decibel gain of the Express?

Both of the other amps you mentioned are around the same at 100 dB. Did they started to oscillating too when you drive the gain knob to the max?

[thinkingchicken]

Just wanna hear a comparison of distortion sounds between high gain tube amps for example those with peak decibel gain beyond 110 dB like JVM410, EVH, Megalith Beta, etc. and high gain solid-state amps, preferably with comparable peak decibel gain.

I think that if you had asked your original question that way, you might have received responses pointing to examples of distortion.

IOW, preferably, solid-state amps that use no clipping diodes at all but just raw gain to achieve clipping.

Why no diode clipping? Is there a particular reason other than what you've read on the internet?

Generally, tube amps sound pleasant while solid-state amps and pedals tend to sound harsh. For example, compare the tube tone of Matthias Jabs from the Scorpions with the solid-state guitar tone of Dimebag.

That is the perceived internet wisdom. Like most internet "wisdom", it is not necessarily so. There has been a great deal of work on making pedals sound not-harsh, and some solid state amps that are difficult to distinguish from tube amps by listening. Can you prove that statement by measurement, or is it only by what you heard on the internet?

Is there a difference between solid-state noise generators and tube noise generators like those from Metasonix? Perhaps, tube-based noise generators have less pronounced high frequencies compare to the solid-state noise generators.

It's difficult to make noise generators of the high-gain type be consistent from one unit to the next, let alone in a different technology. It's even harder to make accurate statements about the general sound of noise generators.

But let's think more philosophically. You're stating a bias against solid state. Can a solid state amp >ever< produce good guitar sound? I've had long discussions on this from people whose mantra was "you can't get tone from a stone". Are you in that camp? Can a tube amp's sound be recorded and sent through a solid state amp and still sound good (assuming it ever did in the first place)? What about a tube preamp through a solid state amplifier that (1) does or (2) does not clip?

Why no diode clipping? Maybe there is a noticeable difference in sound when we're comparing the raw gain between a high gain solid-state amp and a high gain tube amp? I understand there is an exception but I think there will be a generalization between these two.

I remember there are several distortion pedals out there that sound really tube-like, Empress Heavy and Friedman BE OD Deluxe come to mind. That's why I said "generally, solid-state devices sound harsh while tube-based devices tend to sound pleasant". I have seen many several people connecting their amps and pedals through oscilloscopes and the resulting waveforms are similar, depending on their circuits.

So, there's no consistency in saying the difference between high gain solid-state noise generators and high gain tube-based noise generators. Meaning their frequency response will be varied. But I'm interested in listening to those noise generators, solid-state or tube-based, while their peak decibel gain is being measured.

Saying it is good or not is subjective. You can talk about differences without thinking which one is better this one or that one. But usually many people prefer tube-based amp or pedal compare to solid-state just because "I missed the time when people were playing Clapton or Hendrix with a cranked Marshall in the late 60s". But for me, why not having both of them? I like to compare this particular thing to another.

P/s : sorry for this double post. Originally, I just want to edit the previous post and now there are two similar posts. Its much easier typing through the PC.