So I have attached what looks like to be the most popular write up on the web for modifying this amp. The article fails to touch upon a couple of aspects of the circuit I think should be addressed. I would like to start out by saying the amp isn't terrible to begin with but I find it to be a bit on the bright side even after changing out the blue marvel speaker. First I want to address the normal channel and then move on to the high gain channel.
First the article doesn't discuss the RC network R8 and C9 on the normal channel. Also the feedback loop around V1B. I will try and remove C9 and C52 and see what happens......hopefully it will flatten the response out before it gets to the tone stack
Has anyone had any success with modding these amps?
Thoughts and comments welcome.
-Bird
First the article doesn't discuss the RC network R8 and C9 on the normal channel. Also the feedback loop around V1B. I will try and remove C9 and C52 and see what happens......hopefully it will flatten the response out before it gets to the tone stack
Has anyone had any success with modding these amps?
Thoughts and comments welcome.
-Bird
Lots and lots and lots of people have modified these amps. Many successfully and are happy, and as with anything, some folks screw it up.
R8/C9? Not much to discuss. The signal comes from C3 where it branches. R8 and VR1 form a voltage divider. So in general, the signal across VR1 is 250k/(250k+680k) percentage of the C3 signal. C9 bypasses some highs around R8, so it is a brightness cap. Sure, try lifting it.
V1b has a 2.2M feedback resistor R21. That is not a lot of feedback, but you could lift it if you like to see what it changes.
Why would you eliminate C52? That is a 0.0015 across the tone stack. it is essentially a rolloff. You are complaining of brightness, that cap helps remove brightness. Especially when you are in the dirt channel.
It isn't on the schematic, but look at the layout next to V1. See that snaky trace where two comb-like traces mesh into each other? That forms a small capacitor from plate to grid. Just a few picofarads.
R8/C9? Not much to discuss. The signal comes from C3 where it branches. R8 and VR1 form a voltage divider. So in general, the signal across VR1 is 250k/(250k+680k) percentage of the C3 signal. C9 bypasses some highs around R8, so it is a brightness cap. Sure, try lifting it.
V1b has a 2.2M feedback resistor R21. That is not a lot of feedback, but you could lift it if you like to see what it changes.
Why would you eliminate C52? That is a 0.0015 across the tone stack. it is essentially a rolloff. You are complaining of brightness, that cap helps remove brightness. Especially when you are in the dirt channel.
It isn't on the schematic, but look at the layout next to V1. See that snaky trace where two comb-like traces mesh into each other? That forms a small capacitor from plate to grid. Just a few picofarads.
Yes Yes rudimentary stuff. But still a brightness cap nonetheless.
I just want to make the feedback flat.
The bright cap in the preceding stage causes higher order harmonic distortions, that if is disabled might not be there to be rolled off. Either way I am trying to keep things flat.
I will look into this....but if anything wouldn't Mr. Miller say this would roll off the highs?
Roll off highs? Sure but which highs? If you have a 2pf cap, I doubt it will respond in the audio range, but it might serve to clean up any potential Rf interference. Note also the little 39pf cap across the input, also to tame any RF. Guys hate it when their amps pick up local radio stations.
Flat, you mentioned that a time or two. Guitar amps are by their nature anything but flat. people want to know what setting on their tone controls is "flat" and there is none. For flat, plug into the PA system, that is what flat sounds like. So I am not quite sure what your intent is there.
And don't forget this amp is a whole system, not just a clean channel. A lot of design thought went into the dirt channel as well, and some of the things in the circuit are probably more important to that, all the spiky distortion elements from overdrive can be irritating, and so rolling them off helps.
Here is something maybe invisible to most. The FX loop. Play the amp and listen. Now plug a cord from FX send to return. And now listen. That simple little emitter follower driving the send is now in the signal path, and some feel it colors the sound. You may like it or hate it, but it is worth a listen, might come in handy sometime.
Here's one for you, even if not in the dirt channel, look up there at C3. The signal branches through the R8 we talked about, and through C4 to the dirt channel. C4 is a 470pf cap, and with that 1 meg pot to ground it ought to have little or no influence. But now C7 - a rolloff for the dirt channel - comes along. With the dirt volume down, no effect, but when that pot is maxed, then we wind up with two 470pf caps in series across the signal from C3. So it MIGHT have some effect on the highs. Listen carefully to the sound through the clean channel and turn the dirt gain up and down and see if you hear any effect on the highs. You might, maybe.
That the sort of analysis you are looking for?
Flat, you mentioned that a time or two. Guitar amps are by their nature anything but flat. people want to know what setting on their tone controls is "flat" and there is none. For flat, plug into the PA system, that is what flat sounds like. So I am not quite sure what your intent is there.
And don't forget this amp is a whole system, not just a clean channel. A lot of design thought went into the dirt channel as well, and some of the things in the circuit are probably more important to that, all the spiky distortion elements from overdrive can be irritating, and so rolling them off helps.
Here is something maybe invisible to most. The FX loop. Play the amp and listen. Now plug a cord from FX send to return. And now listen. That simple little emitter follower driving the send is now in the signal path, and some feel it colors the sound. You may like it or hate it, but it is worth a listen, might come in handy sometime.
Here's one for you, even if not in the dirt channel, look up there at C3. The signal branches through the R8 we talked about, and through C4 to the dirt channel. C4 is a 470pf cap, and with that 1 meg pot to ground it ought to have little or no influence. But now C7 - a rolloff for the dirt channel - comes along. With the dirt volume down, no effect, but when that pot is maxed, then we wind up with two 470pf caps in series across the signal from C3. So it MIGHT have some effect on the highs. Listen carefully to the sound through the clean channel and turn the dirt gain up and down and see if you hear any effect on the highs. You might, maybe.
That the sort of analysis you are looking for?
Agreed. Between the tone stack and also the speaker..... or I should say the interaction between output and speaker makes for anything but a flat response. Most if not all of the amps I really like have have a flat response up to the tone stack. I do like the common Blackface tone stack with the fixed mid scoop. I am just trying to get in the minds of the people who designed this amp I guess. The input caps were never in question, the .1 and the 470k create a high pass filter setting the lowend response at 3Hz. The 47k and 39pF create a low pass filter getting rid of unwanted highs above 86,871Hz. I guess most things in music is subjective, for example on every amp I have ever played through I have never used the bright switch unless I am using an overly dark sounding guitar, on my Deluxe Reverb I removed the bright cap entirely from channel two because it isn't switchable in that model. So my initial thought when looking at the schematic was get rid of C9. I might not touch C52 but I am thinking after getting rid of C9 I might have too to get the highs back.
Yes with that gain channel 1Meg pot maxed the two caps could be modeled as a 235pF cap to ground surly rolling off the highs, not too sure if it will be audible or not, espcially with a guitar speaker.
Yes
Thank you for your keen eye and knowledge. It's fun to look at amp designs and see what makes them sound the way they do. I was never one on disecting animals in lab......too much of an animal lover but I don't mind disecting amplifiersThanks again Enzo for sharing your thoughts with me.
Last edited:
If it's OK and this thread seems to be done, in dissecting the Classic 30, why do they have a 100 ohm 5 watt screen grid resistor on only two of the EL84s, these being R43 and 53? It's not like it is in parallel to a pair and thus is like having a 200 ohm resistor for each tube of the pair. V5,6 don't have one and V4,7 do. I'm considering adding the other two, but am confused why they have only two.
Hey boobtube, yes it is okay and relevant to discuss R43 and R53, the two large 100R 5W screen grid resistors. I have thought the same thing, why only have these on two tubes? I would think that they would slightly change the operating point of the tubes when compared to the ones with out the screen resistors when driven hard. Not sure how this would translate to how it sounds, probably not much. The other thought I had was they are there for a compramise, the amp sounds better without the screen resitors but is more prone to tube failure.
On another note I was not finished modding the Classic 30. I just got a chance to really decide what I want to do. I also ran some simulations to confirm that what I do will accomplish what I am trying to accomplish. Yes I have added a small value cap from plate to grid on V1A, I can't measure the exact capacitance of this PCB cap but unless it is larger than 10pF it shouldn't make too much of a difference, I guestimated 2pF.....I know highly scientific
First thing I want to try is remove C9 which I just did, now I have to put her back together and give a listen. I want to listen first before I do anything else but I have a feeling I am going to want to remove C52, I am willing to bet this was added to roll off some of the highs created by the treble boost of C9. I am going to leave the tone stack the way it is, for now. This amp doesn't look to be using too much global feedback, maybe 3db. I don't think C37 is there for stability seeing how little amount of feedback is used so I am going to probably end up removing that too, no need in rolling of the highs if they aren't boosted to begin with. These mods will effect the gain channel so it will be dealt with after, besides it doesn't get used anyway, focus is mainly on getting the clean channel where I want it.
It's okay you don't have to start a new thread. I would like to discuss the same resistors so there is no point in having two threads talking about the same amp
You will find the same screen resistor arrangement on a number of other models too. The resistors are there to prevent the two tubes on a side from interacting with each other in undesirable ways, sort of a decoupling resistor if you will. It is like the tubes have no screen resistor at all really. Well sorta anyway.
If you have 5ma of screen current, that 100 ohm resistor only represents a half a volt difference. I am not sure that will make a lot of shift in operating points. But there is nothing wrong with adding two more if you like.
Your meter doesn't go below 10pf? If it has resolution, try this. Ever weigh a small dog or cat that won't sit still? The trick is to weigh yourself, then weigh yourself holding the pet, then subtract your weight from that. You are left with the pet weight. SO you want to measure what might be 2pf? Tack something like a 22pf or whatever small one you CAN measure in parallel, measure that combination, then measure your testing cap alone. The difference in readings will be the capacitance of that board thingie. However, since it is such a small value, you may have to disconnect the other parts from it so they do not form parallel paths. I see four components to lift, if you do it.
If you have not worked on C30s a lot, I URGE you to keep the board flexing to a minimum. The one board to board junction has a couple little L brackets holding them in position, so those jumpers never bend, but the other board to board junction is the one you bend, and it only takes so many flexes before those jumpers start to crack. learn to flex it just enough to clear going in and out, and only open it up the minimum to get at the solder. And even if you don't think you cracked any of the wires, once it is in place, give a little tug to each wire down the row just to make sure.
Steve's Blue Guitar site has a bunch of various mods, so you may like parts of one and parts of another. I even had a hand in one. For a customer, we converted it to cathode bias, and just for the why not, I added a switch, so you could run it either way, cathode or fixed. It isn't a huge change, not like something you'd change song to song, chances are you'd set it one way and leave it. But it did change the dynamics of the amp. Since the cathode bias I chose wound up hotter than the stock fixed, I altered his amp for adjustable bias, so we could set the fixed to a comparable level. Just doing that bias mod made a difference in the amp.
yes we added a bias test point with a 1 ohm resistor. Took a little judicious trace cutting, but worked out well and conveniently.
If you have 5ma of screen current, that 100 ohm resistor only represents a half a volt difference. I am not sure that will make a lot of shift in operating points. But there is nothing wrong with adding two more if you like.
Your meter doesn't go below 10pf? If it has resolution, try this. Ever weigh a small dog or cat that won't sit still? The trick is to weigh yourself, then weigh yourself holding the pet, then subtract your weight from that. You are left with the pet weight. SO you want to measure what might be 2pf? Tack something like a 22pf or whatever small one you CAN measure in parallel, measure that combination, then measure your testing cap alone. The difference in readings will be the capacitance of that board thingie. However, since it is such a small value, you may have to disconnect the other parts from it so they do not form parallel paths. I see four components to lift, if you do it.
If you have not worked on C30s a lot, I URGE you to keep the board flexing to a minimum. The one board to board junction has a couple little L brackets holding them in position, so those jumpers never bend, but the other board to board junction is the one you bend, and it only takes so many flexes before those jumpers start to crack. learn to flex it just enough to clear going in and out, and only open it up the minimum to get at the solder. And even if you don't think you cracked any of the wires, once it is in place, give a little tug to each wire down the row just to make sure.
Steve's Blue Guitar site has a bunch of various mods, so you may like parts of one and parts of another. I even had a hand in one. For a customer, we converted it to cathode bias, and just for the why not, I added a switch, so you could run it either way, cathode or fixed. It isn't a huge change, not like something you'd change song to song, chances are you'd set it one way and leave it. But it did change the dynamics of the amp. Since the cathode bias I chose wound up hotter than the stock fixed, I altered his amp for adjustable bias, so we could set the fixed to a comparable level. Just doing that bias mod made a difference in the amp.
yes we added a bias test point with a 1 ohm resistor. Took a little judicious trace cutting, but worked out well and conveniently.
Current setup has yielded the best results so far, C9 and C37 have been removed. I tried removing C52 and the amp sounded too aggressive in the upper mids and highs along with making the high gain channel sounding too fizzy so it got put back in. I also added a 500pF polystyrene cap in parallel with C4 and this helped to thicken up the high gain channel. I think I will leave this amp the way it is for now and then maybe mess around with the tone stack, for now the amp has a much more pleasing sound to my ear in the upper mids and highs.
Enzo, you sure do know Peaveys!
Famousmockingbird like Enzo says, remember this is a system.
IMHO they purposely boost highs radically before they generate distortion, then allow you to cut them somewhat after. The basic guitar sound gets treble-boosted then can be cut in the tone stack, which is relatively late in the signal chain so you can roll off the irritating most buzzy treble of the distortion without losing the sparkly treble of the basic guitar sound. So indeed, the first stages are not voiced great for a baritone guitar, big semi-hollow jazz, fender kind of full scooped-mid sound, much less a bass guitar. Indeed treble boost is achieved via gain and rather sharp bass cut. But remember that when you subdue that treble boost that's added before the distortion is generated, it means that the distortion sound and the guitar sound will be more similar to each other, and you will no longer be able to cut the treble in the tone stack or later stages as much to make the distortion more 'creamy' and less irritating. If you like the way the distortion emphasizes the lower beats in a chord, you might lose some of that too. If you don't boost the treble on the guitar early, it not only changes the harmonic content of the distortion generated, it prevents you from cutting the treble on the combined guitar & distortion signal later. In attempting to get the whole amp 'flatter' and the guitar sound less "trebly" you might end up with the distortion much more irritating and trebly and lose some of the independent control of distortion. It seems you already know that the speaker is the final filter...
One aspect of several is to think of this like applying a pre-emphasis curve and and then a correction curve with knobs. Like the pre-emphasis before tape, and correction after. Or the recording curve before they cut a record master, and the RIAA curve applied afterward. There is a curve applied before distortion is added, then another after. The one before changes the nature of the distortion generated, the one after changes the tone of the distortion. Done carefully, the basic guitar may come thru the system with only subtle changes compared to the radical changes it underwent thru sections of the signal chain.
If you try to 'voice' the amp like a hi-fi, the distortion characteristics may not be what you wanted. Think carefully not only of the change you want, but about what stage you want it at. You have to voice the front-end and back-end separately but as a system. REMEMBER, THE GUITAR GOES THRU BOTH THE FRONT-END AND THE BACK END. BUT THE DISTORTION GENERATED IN-BETWEEN ONLY GOES THRU THE BACK END. You may even inject irritating clipped test signal before the tone stack to voice the back, then when you have the distortion voiced the way you want it, go back and inject cleaner guitar at the input and voice the front for the system sound. That's backward from how most people approach it their first times. And that doesn't take into account how the distortion itself is different when generated on the input that has pre-emphasis applied.
Most of all, don't just listen clean. Play clean, dirty...
And try to use switches which make your changes reversible by the user.
I recommend you alter the slopes of the treble boost at the front end rather than just their hinge points. Sometimes that may mean altering the bypass resistors instead of just the cap values. That keeps more of the lovely distortion characteristics of these amps yet gives you more bandwidth thru the system. Once treble or bass is completely removed it's gone for good, so you might want to keep more material intact. The original voicing is as much about voicing the distortion as voicing the guitar...maybe more. That's how an amp distinguishes itself in the marketplace.
Keep in mind I don't have Enzo's experience, and I experiment more with the related Classic 50 and Classic 100, and don't have experience with the 30. Have fun!
Famousmockingbird like Enzo says, remember this is a system.
IMHO they purposely boost highs radically before they generate distortion, then allow you to cut them somewhat after. The basic guitar sound gets treble-boosted then can be cut in the tone stack, which is relatively late in the signal chain so you can roll off the irritating most buzzy treble of the distortion without losing the sparkly treble of the basic guitar sound. So indeed, the first stages are not voiced great for a baritone guitar, big semi-hollow jazz, fender kind of full scooped-mid sound, much less a bass guitar. Indeed treble boost is achieved via gain and rather sharp bass cut. But remember that when you subdue that treble boost that's added before the distortion is generated, it means that the distortion sound and the guitar sound will be more similar to each other, and you will no longer be able to cut the treble in the tone stack or later stages as much to make the distortion more 'creamy' and less irritating. If you like the way the distortion emphasizes the lower beats in a chord, you might lose some of that too. If you don't boost the treble on the guitar early, it not only changes the harmonic content of the distortion generated, it prevents you from cutting the treble on the combined guitar & distortion signal later. In attempting to get the whole amp 'flatter' and the guitar sound less "trebly" you might end up with the distortion much more irritating and trebly and lose some of the independent control of distortion. It seems you already know that the speaker is the final filter...
One aspect of several is to think of this like applying a pre-emphasis curve and and then a correction curve with knobs. Like the pre-emphasis before tape, and correction after. Or the recording curve before they cut a record master, and the RIAA curve applied afterward. There is a curve applied before distortion is added, then another after. The one before changes the nature of the distortion generated, the one after changes the tone of the distortion. Done carefully, the basic guitar may come thru the system with only subtle changes compared to the radical changes it underwent thru sections of the signal chain.
If you try to 'voice' the amp like a hi-fi, the distortion characteristics may not be what you wanted. Think carefully not only of the change you want, but about what stage you want it at. You have to voice the front-end and back-end separately but as a system. REMEMBER, THE GUITAR GOES THRU BOTH THE FRONT-END AND THE BACK END. BUT THE DISTORTION GENERATED IN-BETWEEN ONLY GOES THRU THE BACK END. You may even inject irritating clipped test signal before the tone stack to voice the back, then when you have the distortion voiced the way you want it, go back and inject cleaner guitar at the input and voice the front for the system sound. That's backward from how most people approach it their first times. And that doesn't take into account how the distortion itself is different when generated on the input that has pre-emphasis applied.
Most of all, don't just listen clean. Play clean, dirty...
And try to use switches which make your changes reversible by the user.
I recommend you alter the slopes of the treble boost at the front end rather than just their hinge points. Sometimes that may mean altering the bypass resistors instead of just the cap values. That keeps more of the lovely distortion characteristics of these amps yet gives you more bandwidth thru the system. Once treble or bass is completely removed it's gone for good, so you might want to keep more material intact. The original voicing is as much about voicing the distortion as voicing the guitar...maybe more. That's how an amp distinguishes itself in the marketplace.
Keep in mind I don't have Enzo's experience, and I experiment more with the related Classic 50 and Classic 100, and don't have experience with the 30. Have fun!
Last edited:
Great explenation and thank you for sharing.
I think some people are more susceptible to being irritated by higher frequencies because they can hear them better then others, I am one of these people. I understand that "good" tone is subjective and the amp manufacturers need to voice their products to set them apart, distortion shaping is the name of the game inside this little amp I have read on other forums that where just replacing the speaker in these amps won't tame the ice pick in the ear effect, again some people might not even notice it if their hearing isn't that great in the higher domain. So these post filters whether in the tone stack or the speaker itself aren't enough to tame what was going on in the front end, only fix was to roll the treble way down and that just doesn't sound good. The guitar passive pickup naturally has a spike in the upper mids region due to the resonance of the inductance and capacitance, this is why I think the mid scoop sound works so well in guitar amps and pedals. This hump along with the fact that guitar amps output impedance being rather high (acting similarly like a current source) the speaker's impedance rises with frequency further emphasizing the top end spectrum, more voltage across a higher impedance. I guess what I am saying is that these boosts are enough for me, I don't need filters to boost the highs in an amp it makes my ears bleed.
I think some people are more susceptible to being irritated by higher frequencies because they can hear them better then others, I am one of these people. I understand that "good" tone is subjective and the amp manufacturers need to voice their products to set them apart, distortion shaping is the name of the game inside this little amp
I have read on other forums that where just replacing the speaker in these amps won't tame the ice pick in the ear effect, again some people might not even notice it if their hearing isn't that great in the higher domain. So these post filters whether in the tone stack or the speaker itself aren't enough to tame what was going on in the front end, only fix was to roll the treble way down and that just doesn't sound good. The guitar passive pickup naturally has a spike in the upper mids region due to the resonance of the inductance and capacitance, this is why I think the mid scoop sound works so well in guitar amps and pedals. This hump along with the fact that guitar amps output impedance being rather high (acting similarly like a current source) the speaker's impedance rises with frequency further emphasizing the top end spectrum, more voltage across a higher impedance. I guess what I am saying is that these boosts are enough for me, I don't need filters to boost the highs in an amp it makes my ears bleed.
Last edited:
Highs are more piercing. 100 watts of 100Hz is a lot easier to listen to than 2 watts of 1000Hz. To me.
One thing a lot of people forget or don't think of is that a hifi amp is designed to faithfully REproduce some sound. But a guitar amp is a part of the instrument, it is a primary producer of sound, not a reproducer.
Botom end is often kept out of overdrive channels because those lows tend to turn into mud. Same with reverb, we really do not want full range sound going through reverb.
One thing a lot of people forget or don't think of is that a hifi amp is designed to faithfully REproduce some sound. But a guitar amp is a part of the instrument, it is a primary producer of sound, not a reproducer.
Botom end is often kept out of overdrive channels because those lows tend to turn into mud. Same with reverb, we really do not want full range sound going through reverb.
The AC50 especially did want to distinguish itself as an EL84 amp, with the 'bright' 'chime' of an AC30 top-boost.
The only important point in my blather is that it does matter where you put the filters, early (on just the guitar) or late (on the guitar and distortion).
BTW I do detect changes in what I find irritating as I age. Some mornings the sound of the bathtub filling seems so loud it seems distorted and painful. Of course it's my ears. Now I know what my dad was complaining about and why he would turn off the circuit breaker when we played electric guitar.
The only important point in my blather is that it does matter where you put the filters, early (on just the guitar) or late (on the guitar and distortion).
BTW I do detect changes in what I find irritating as I age. Some mornings the sound of the bathtub filling seems so loud it seems distorted and painful. Of course it's my ears. Now I know what my dad was complaining about and why he would turn off the circuit breaker when we played electric guitar.
Typo, i meant "The Classic 50 especially did want to distinguish itself as an EL84 amp."
The AC50 was an EL34 amp.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.