Wanna help with basic mods...

[isaac]

I've been kinda bored lately, so I thought that I'd post this up here...

I always love trying to figure out how to tweak amps, and helping others, so if you have a basic idea for a mod that you like, ask me, maybe I can help a little. Make sure to note the amp, and post a schematic.

[Tone Meister]

It seems like I'm always tinkering with the Hot Rod Deville and Hot Rod Deluxe amps, looking for ways to increase the headroom in the clean channel. No matter which revision of either amp, I will nearly always default to the III series schematics, which are attached. I'm well acquainted with speaker and power tubes upgrades and their impact on the overall tone.

We don't have to limit the discussion to the clean channel, as I'm open to any sensible circuit mods that are verifiable to be actual improvements over stock.

[soundguruman]

The only way to increase the headroom is to increase the available voltage.

[isaac]

True ^^ I personally have not worked on a HRD. What you could do, is change the speaker to a more efficient speaker. And for a tighter and maybe more perceived headroom (if you understand what i mean by that) you could lower the value of the NFB resistor. That might make it a little more bright than you would like though, so you could also drain more treble to ground, but there is a certain point where you were better off before because you could have changed SO much stuff.

[R.G.]

... I'm open to any sensible circuit mods that are verifiable to be actual improvements over stock.

Quoting the bard, aye, there's the rub. That word "verifiable".

Altogether too many people think that if they hear it, it's there. And this has been exhaustively proven to be unreliable by decades of psychological and psycho-acoustic work on how the human mind decides things. Worse yet, the worn-out battle cry of the hifi tweako fringe that no only can THEY hear magic differences, but that anyone that doesn't hear these purported differences is deficient in either hearing or mental acuity. It's the classic "if you disagree with me, you're not only wrong, you're incompetent in some way" spin.

Decades of research into human perception has led to the view that if you can't even theoretically measure it without the measurement relying primarily on human perception, it's highly likely that it's self-deception at work. Therefore the difficulty of how you'd "verify". Voting isn't good enough, nor is the testimony of self-proclaimed "experts".

The only way to increase the headroom is to increase the available voltage.

Well, there is always lowering gain. If you have a stage in the middle that's clipping, lowering that stage's gain gives more real headroom without lowering the amp's output at all. Careful thought about the gain and signal levels at each stage of an amplifier makes the most headroom for the available power supply.

[isaac]

Well, there is always lowering gain. If you have a stage in the middle that's clipping, lowering that stage's gain gives more real headroom without lowering the amp's output at all. Careful thought about the gain and signal levels at each stage of an amplifier makes the most headroom for the available power supply.

That was my other thought, say if an amp had 3 or 4 gain stages in a clean channel (honestly... that's not usual, but for an example...) then you could reduce it to 2 stages or so, that would give more "headroom."

[Tone Meister]

Quoting the bard, aye, there's the rub. That word "verifiable".

Altogether too many people think that if they hear it, it's there. And this has been exhaustively proven to be unreliable by decades of psychological and psycho-acoustic work on how the human mind decides things. Worse yet, the worn-out battle cry of the hifi tweako fringe that no only can THEY hear magic differences, but that anyone that doesn't hear these purported differences is deficient in either hearing or mental acuity. It's the classic "if you disagree with me, you're not only wrong, you're incompetent in some way" spin.

Decades of research into human perception has led to the view that if you can't even theoretically measure it without the measurement relying primarily on human perception, it's highly likely that it's self-deception at work. Therefore the difficulty of how you'd "verify". Voting isn't good enough, nor is the testimony of self-proclaimed "experts".

We are absolutely on the same page here RG, hence the reason I purposely used the word verifiable. I can assure you that if the end result isn't quantifiable, I would have no problem rejecting that "mod."

Well, there is always lowering gain. If you have a stage in the middle that's clipping, lowering that stage's gain gives more real headroom without lowering the amp's output at all. Careful thought about the gain and signal levels at each stage of an amplifier makes the most headroom for the available power supply.

^^^^^ This is more in line with what I am thinking. I already know I can install a 100+ dB speaker and 7581 power tubes that draw lots of current and get the power amp section where I like it - I have verified that headroom increased by using the preamp of a second amp to drive just the power section of the amp using the power amp input. Before and after, of course.

So specifically, I'd like to take a closer look at the various stages of the preamp circuit to see if the gain can be more evenly distributed, the goal being to end up with a cleaner clean channel and have the amp become progressively more aggressive as the Drive and More Drive "channels" (circuits) are accessed.

That was my other thought, say if an amp had 3 or 4 gain stages in a clean channel (honestly... that's not usual, but for an example...) then you could reduce it to 2 stages or so, that would give more "headroom."

I've experimented with the preamp some, very carefully, and have had some reasonable results. But I'm sure some of you engineer types can analyze the preamp and come up with some ideas that end up viable. I saw this thread and couldn't resist testing the waters!

[isaac]

I started the thread basically for the basic mods etc, but I'll try a little on this one...

One thing I see right off the bat with the HRD's is the mid pot... I'd suggest replacing that with a 10kL and putting a jumper from the middle to the top, to be like the twin tonestack.

One thing you could try is a 1M volume pot, instead of the 250k. Or at least a 500k. I'm not sure why they picked the 250k, but I'd at least try a 1M. I'm assuming the 250K value pairs with teh 240k resistor right before it, so at the highest position, it is like a voltage divider (even more than a regular volume pot) to keep the level down. So increasing the volume pot value up, might increase gain, but it could also increase loudness, and wake it up a little bit. And PLEASE, if anyone disagrees, please butt in.

Now, I've also heard that the HRD is a bit muddy, but again, never tried one. So you could try this, again, these are theoretical... C9 is a 22uf, you could try a 5uf or a .068uf. Lower C10, maybe try a .01. C11 is a cap to ground, nt adjustable... That one could be lowered to bring back highs that weren't there, or just removed. Also, if the whole amp could use some highs, (i don't know, just thinking out loud) you could remove C28 and C24 could be lowered for less low end. And one thing I do in EVERY amp pretty much, is change what is C26, and 27 in this amp, I make those .01, or possibly .022. Usually tightens up the low end, and just sounds sweeter. That's IMO. But some of these are in general, some are ideas for this or that.

And once more, I've never tried one of these, or fixed/modded one either, so remember that when you look at these.

[R.G.]

"Headroom" is tossed around nearly as much as "tone", and has almost as much meaning in the general use. It does my heart good to see how you're using it, along with "verified".

For a single stage, "headroom" is tied up in available output swing, as determined by the output device(s) and circuit, the circuit gain, and the input circuit.

I think what SGM was trying to get at was that a given power supply voltage and output circuit configuration can only swing as much as the power supply **and power devices** and ** output circuit** will let it, no matter what you do to the input. That is true, if limited.

The output voltage available from a tube amp output stage is limited by the available B+, the output tube type, which limits how close to ground the tubes can pull the OT primaries, by the transformer ratio, and by the loading on the circuit. Once those are chosen, you get a max output swing for that set of conditions.

But how much input signal can you put into the tubes and have them swing that maximum? That depends on the stage gain, as the output is loaded, and on the input compliance of the tube control grid. Each tube has an input limit. It can swing from the grid cutoff voltage on the grid to Vgk = 0 before grid conduction starts, distorting the output. If you have a low-impedance follower driving the grid, the grid conduction changeover can be much less abrupt, as the plate will obligingly continue to conduct more current for Vgk>0 for a while. But there are limits. In a single tube setup, the range of cutoff to Vgk=0 is all the INPUT headroom you have. In a fixed bias push-pull amp, the input headroom is 2*Vbias before you get grid limiting. Notice that headroom is also dependent on the bias point. The more "off" a push-pull stage is biased, the more input voltage it can swing before grid clipping.

Grid limiting happens whether the output stage is swinging all it can or not. And you reach a max headroom point where the grid swing just before grid clipping is just enough when multiplied by the tube gain to swing the output plates all they can before they run out of either current drive ability or voltage swing. If all three aren't right, the headroom goes down.

That's for the output stage. Each stage all the way back to the input has a similar balance to be reached for maximum headroom on the entire amp. A P-P output stage might not be able to take more than +/-40V peaks before grid clipping happens. If the OT is well chosen and designed to reflect the load well, the output tubes will just be hitting cutoff and/or current limit as the grids hit clipping. The PI stage doesn't really need to put more than +/-40V peaks out, then, if you're after headroom, not clipping. So the PI stage and its bias point, loading, and gain set a limit on how big an input it can take without driving the plates into saturation and/or cutofff. And the input of the typically duotriode 12AX7 in the PI can't usually take more than about +/-1.5V peaks before IT goes into grid clipping. So the PI really only needs a gain of about 27 to drive 1.5V peak signals to 40V peak signals to just drive the power amp to the edge of clipping.

And so on back to the input. Each stage has a max input based on input clipping, gain, and loading as well as power supply. If you tune each one, you can reach a point where the entire amp is running just at the edge of clipping for the maximum input signal you're likely to get.

When this has happened, you've reached the maximum headroom achievable for that power supply voltage for those devices in those circuits. If any of the stages clips early, it doesn't matter what headroom is available after that - the later stages very linearly preserve the already clipped signal, which is likely not to be much help if you want headroom.

But that's not how modern amps are designed. Modern MV and channel switching design deliberately clips before the output stage and then attenuates at the output stage to tame things back down.

In the one amp I tuned for maximum headroom, I did an analysis of the gains and found that I could get the amp to just distort where I wanted by putting a resistive attenuator after the first stage. The first stage was run full out, to get the signal up out of the noise, then the signal was attenuated down to an intermediate level, put through the T-V stacks and a post amp, then into PI and output. Juggling the fixed attenuation let me have enough gain to play single coils so they were just barely sending one or another stage into the edges of clipping (component variations forced this to be somewhat indeterminate) as the volume control was turned up to about 9. I could literally adjust the range before clipping with two resistors, so it could turn into a distortion preamp, or back down to a hardly-distorts-at-10 amp.

It takes some work and understanding of what each stage does.

[Tone Meister]

"Headroom" is tossed around nearly as much as "tone", and has almost as much meaning in the general use. It does my heart good to see how you're using it, along with "verified". ...

While I do not understand headroom within circuits at any real depth, I do have a reasonable grasp of the general concept as an audio professional. Headroom is one of the primary goals when we rack up twenty MacroTech 5000s and another twenty MaroTech 3600s in a touring rig. Amplifiers like that require power, and we provide that power by stringing 600 amps X 2 of AC via minimal runs of 4/0 SJOOW cable to two power distribution racks. Lighting is another matter, and the lighting rigs have their own distribution and another 600 amps of AC on separate runs of that same 4/0 SJOOW cable, which draws its power from a separate sub panel or generator.

But we're talking headroom in a guitar amp and its effect on tone, one of my favorite subjects.

... It takes some work and understanding of what each stage does.

Busted! That is certainly much more answer than I am prepared to digest, but you already knew that. I'm definitely not looking to absolutely maximize headroom throughout the entire amplifier because that certainly isn't practical. Plus, it doesn't help that I do not fully grasp what each stage does. But you did hit on a point on which I think best illustrates what I'd like to do, not only in the Fender Hot Rod amps but in other amps that are similarly designed. You mentioned "Modern MV and channel switching design deliberately clips before the output stage and then attenuates at the output stage to tame things back down." It is this deliberate clipping and subsequent attenuation that I'd like to understand and learn to adjust in a repeatable way.

I have learned that upgrading to a very efficient speaker and higher wattage power tubes make a marked headroom increase in the power amp - that is something I have verified for myself and am satisfied with the results. Now, I'd like to explore adjusting this deliberate clipping and attenuation while operating within the voltage constraints of the amplifier(s).

[R.G.]

While I do not understand headroom within circuits at any real depth, I do have a reasonable grasp of the general concept as an audio professional. Headroom is one of the primary goals...But we're talking headroom in a guitar amp and its effect on tone, one of my favorite subjects.

All same thing really. "Headroom" gets tossed around in the fact-free press releases of audio hifi tweako, and increasingly in musical equipment advertising like it means "how much *more* you can push it without it breaking", or some such. It's simpler than that. In a house or car, "headroom" means "how tall can you be before you bang your head?".

Same here. Headroom just means "how big a signal can you put through it without distortion?" In an audio amp where the signal must go through every stage to wind up at the output, it's like a long hall way with varying ceiling heights. Since audio can't bend over and then stand up (without companders, that is! ) the lowest ceiling determines the headroom for the passageway. In an amplifier with multiple stages, "maximum headroom" ultimately becomes "all stages start distorting at the same time."

Busted! That is certainly much more answer than I am prepared to digest, but you already knew that.

Actually, I didn't know you were busted You were a lot more coherent that some of the "experts" we get here.

I'm definitely not looking to absolutely maximize headroom throughout the entire amplifier because that certainly isn't practical. Plus, it doesn't help that I do not fully grasp what each stage does.

Yeah, maximum headroom requires some careful tinkering with everything along the way. However, if you identify a headroom bottleneck (to profoundly mix metaphors ), you can increase headroom for the whole amp by relieving that one bottleneck. Surprisingly, computers do much the same for performance. Computers can have incredibly fast CPUs but be "memory bound" by not being able to get data to and from the CPU fast enough; or "IO bound" by not being to load/store data fast enough; or "compute bound" by having plenty of memory bandwidth and I/O, but not being able to slice and dice it fast enough. Relieve the bottleneck, and the whole thing gets faster.

But you did hit on a point on which I think best illustrates what I'd like to do, not only in the Fender Hot Rod amps but in other amps that are similarly designed. You mentioned "Modern MV and channel switching design deliberately clips before the output stage and then attenuates at the output stage to tame things back down." It is this deliberate clipping and subsequent attenuation that I'd like to understand and learn to adjust in a repeatable way.

I have learned that upgrading to a very efficient speaker and higher wattage power tubes make a marked headroom increase in the power amp - that is something I have verified for myself and am satisfied with the results. Now, I'd like to explore adjusting this deliberate clipping and attenuation while operating within the voltage constraints of the amplifier(s).

You already know some of the answers, you just don't know you know them. You're accustomed to dealing with issues of headroom on the pro stage. Ultimately, the performer's guitar (if it's a guitar) doesn't get heard directly by the audience. It comes out of those behemoth speakers for the PA. The performer may be getting a great sound out of his amp on stage, but the audience won't hear it until you mike it up. The performer's guitar and amp create a distortion sound, but you take that and condition it to drive the PA speakers in a way that the PA makes the sound louder without changing it noticeably.

A MV amp is a microcosm of that. If you are using preamp stage clipping to generate the distortion, you have to then get that heard. You do that by filtering for tone, then amplifying it up to drive the later stages and finally the speakers. In a guitar amp, the power amp and speakers necessarily contribute some tone shaping of their own, but to get the maximum output levels from an output stage and speakers, you need to (a) know the output level they can handle without hashing it up on their own and (b) feed the earlier, carefully-distorted stuff to them at a level they can use right.

If the earlier distortion only provides a peak signal that's half of what the power amp needs for full power - yep, it ain't gonna put out full power. If the earlier distortion provides a signal to the power amp that is four times as big as it can handle without clipping, you're going to be listening to the clipping of the power amp, not the earlier, carefully-done preamp distortion. You might even like the "touch sensitivity" you get from the switchover from one stage's distortion to another... oops! I let out a secret, didn't I?

So think of it that way. If the power amp itself is not being driven into clipping, when you're hearing distortion it's another stage doing it. Which stage? And how does that stage clip? That's where the art is buried.