Please help with 2ch 50w build - Schematic supplied

[Meat_And_Beer]

Try this ...................  use two 470k mixing resistors as a way to mix those two channels and bypass the "mixing tube".
See if that resolves the problem.  IF it does, then the issue is the mixing tube.

I didn't have any mixing resistors, so I just tried the regular kind... 

It didn't help us, but I thought that was an excellent idea, thanks for the suggestion!

That figures, about the mixer ckt being alien to you guys. This was one of those things where you have an idea, ponder it, and when you wake up the next day the answer pops up. (Origin of this ckt)

I believe that is what I saw in the tube manual, I was just flipping through the different schems one day and remembered seeing it. And it would also figure it's nothing like this one, what do I know?

I keep trying to remember why exactly we tacked that bypass cap and additional r on the cathode resistor... But I can not, and it was a while ago.

Great, we've made some progress. This is a pretty sure sign that there is something wrong in either FX, PI, PA or PS.

I have tried jumping straight from the mixer to the PI and pulling the FX tube, no change. We also have solidified the PA and PI as known good. I swapped every single filter and decoupling cap on the supply with known good ones, one at a time. Nothing changed there either. Good thought though, for sure. It took me a LONG time to convince myself that the FX loop wasn't the issue. I've proved it's kosher time and time again, but I still glare at it suspiciously on occasion. Well, I do that to a lot of things in there now...

I'm going to try to pull power supplies to all stages of channel 2 and see where that leaves channel 1. If that doesn't help, I'll try the other.

[Meat_And_Beer]

I was looking at your photos and with the amount of shielded wire you are using you could have easily melted the insulation for the inside conductor and it's shorting out to the shield. It has happened before to the best of them.

Does the shield have clear insulation on it or is it bare? If it's bare it could have shorted out something.


                  Brad       

I thought this was right up my alley. I poked around for continuity to the chassis where there really should be none... and I think you're on to something.

The shield is bare, and where it gets close to things I don't want it to touch I shrink tube it, or solder it somewhere for stability. I also tried to supply all of them, as one, to ONE main entry to the buss bar.

Is anyone thinking ground loop problems with this?

[John]

I was looking at your photos and with the amount of shielded wire you are using you could have easily melted the insulation for the inside conductor and it's shorting out to the shield. It has happened before to the best of them.

Does the shield have clear insulation on it or is it bare? If it's bare it could have shorted out something.


                  Brad        

I thought this was right up my alley. I poked around for continuity to the chassis where there really should be none... and I think you're on to something.

The shield is bare, and where it gets close to things I don't want it to touch I shrink tube it, or solder it somewhere for stability. I also tried to supply all of them, as one, to ONE main entry to the buss bar.

Is anyone thinking ground loop problems with this?

Possibly, although the amp has worked before. That is an awful lot of bare shielding which means a lot of opportunity for something to go wrong/start touching. I'm 100% positive your amp friend has forgotten more than I'll ever know, but I'm surprised he let you use bare shielded wire. I know just enough to know I could never get away with it. ;) Those are also an awful lot of pots and switches to go dodgy or pick up stray noise.  

Oh, and dude... your avatar is freaking me out. *shudder*

[Meat_And_Beer]

Better? I guess sheetrock screw skull head guy is pretty aggressive looking. He was smiling though. 

My amp guy was the one who suggested this wire. I don't think he quite anticipated me using this much of it though...

The amp DID work before. Since then, my hands have been slapping the inside of this thing silly, anything could have happened by now.


Hey, guess what I have... A stupid question. I try to limit these, and find out the answer myself elsewhere, but I cant seem to get a solid yes or no.

Should my OT primaries have continuity to chassis? Of course the common should, but if so, I completely don't get how that would even work..

Not quitting the day job any time soon.

[HotBluePlates]

Should my OT primaries have continuity to chassis?

No.

The OT primary winding has a center-tap that should go to a high-voltage point in your B+ rail. The ends of the primary winding go to either side of the push-pull output stage.

If there was continuity to the chassis, then you'd either be shorting the B+ rail to ground (if the chassis is firmly bonded to ground), or placing high voltage on the metal chassis (if the chassis is not firmly bonded to ground). Either one is bad, for different reasons.

[Willabe]

Oh, and dude... your avatar is freaking me out. *shudder*

Thank you John.

Better?

Yes.


               Brad     

[Meat_And_Beer]

Should my OT primaries have continuity to chassis?

No.

The OT primary winding has a center-tap that should go to a high-voltage point in your B+ rail. The ends of the primary winding go to either side of the push-pull output stage.

Good, it does all that. I'm stupid and asked the wrong question then. 

It's the multi-taps out to speakers. 4 ohm, 8 ohm, etc. Why would there be the common for the - signal, and the same common be riding on the + signal to the speakers??

[HotBluePlates]

Your OT primary may be 100-200Ω typically from center-tap to either end. The OT secondary from common to 16Ω tap might only be a few ohms. Therefore, you will get continuity from the highest impedance tap to the chassis, where the common is grounded.

"Continuity" for most meters does not mean "0Ω" but maybe something more like "less than 100Ω". Or whatever resistance the manufacturer decided represented "substantially connected."

[zendragon63]

Interesting creation meat and beer. I can appreciate that you are chasing the tone you ultimately want. And that you started out with circuits building blocks that others designed (i.e. Engl and BK Butler). And can appreciate you have given your word on the schematic thing. However...(FWIW), unless a 'guru' goes by the initials HAD or is or planning to market this secret sauce circuit to the world for livelihood, it doesn't make a lot of sense to horde good knowledge--eventually it will get out and the guru will likely be forever known as selfish/self-serving. Honorable grandmasters of tube amplification shared their 'secrets' with the those worthy of the knowledge (though I could be wrong and often am) and I know of no members of this forum that have built empires from any secret sauces shared freely here. But everyone has their reasons...so I will not judge but only offer a thought.

I have read and reread this thread several times and offer another thought--since your amp worked up to the time of the final FX modification, logically it makes sense to retrace everything subsequent to that mod. BUT since you have been in and out of the chassis as many times as you indicate, and no progress is being made yet, I humbly suggest that you take an afternoon or evening and thoroughly visually check EVERY SINGLE solder joint and reflow ANY that even look remotely suspicious--you will probably have to cut most of your cable ties to get down a level or two. I have had at least two builds in recent memory that cursory inspection looked good and the amp worked but 'not quite right' and there was a fluctuating measurement and sound. The culprit was only as simple as a solder joint. Again, even if you are experienced wielding a solder iron, there are a lot of connections here in your build so the potential is there.

Again, for a relatively new amp builder, nice effort! Regards

dennis

[Meat_And_Beer]

Thanks Dennis!

Of course, I have found one cold solder joint about 2 weeks ago. Was that the only one? Ppphhhsssssshhhhh.. Probably not!

The things that are iffy to me, is when I rotate the 1st stage tube in its socket for either channel, a LOT of fluctuation/noise comes about. Sometimes it'll sit there sounding cruddy, and I'll poke it so it sits in a microscopically different position, and the cruddy-ness disappears. I have re-flowed these solder joints, and I have cleaned all pin holders with deoxit. I cant get in there to tension them though... They were new when I purchased them at the beginning of the build. Make no difference their age? Thoughts on that?

I tried Willabe's idea concerning all the shielded madness. I found no continuities to ground on the center wire. Excellent idea, I was hopeful on that one!

I removed power supplies for each channel and their tubes. No effect removing channel 2 and listening to channel 1... The other way around seems to make a big improvement, but it was still there. And the issue of it's (Channel 2) second tone networks treble was still acting up. (At zero, its fine, as you bring it up, treble increases, but volume decreases.)

hhoooooo boy.

So I'm finishing up the updates & cleaning up on a very large schematic in photoshop. I'll plunk her up here after dinner tonight.

It is very unfortunate I have left a sour taste to many people by not releasing it. I was trying to first see if anything could come from my very lengthy information intro, and subsequent tests. I'm hoping after it's up later, all the animosity will disappear. Just trying to be a man of my word, I know most of you can respect that. And I do trust you guys, I've been following for a while, I see your tight knit community! It's just that when you put something on the internet, even momentarily, it is forever on the internet.


Chris

[Katie 77]

actually, the Sound City L/B100 mixes the two 1st stage preamps in a VERY similar fashion

[Meat_And_Beer]

actually, the Sound City L/B100 mixes the two 1st stage preamps in a VERY similar fashion

You mean the input jack going to both pre's at once? I tried a pickup selector at first to switch one, the other, or both. But there was too much capacitive coupling in the switch because of sheer proximity so they were both always getting signal for the most part. Shunting the signal at the end works fabulously.

Wish I never got rid of my Sound City. I was a kid, and I thought it was just too loud! Silly kids. Practically gave it away too...

[Meat_And_Beer]

So, a pretty interesting thing just unfolded here at the broken amp camp.

I was plucking the twanger (On gear that works..) and a chord was ringing out into silence. In the very lowness of it, before my noise gate stood in the way and put up its wall, my head snapped toward the speaker cabinet...

...The "ground swirl"...

It's my friggen guitar! I only have 5 of them, could I possibly be bothered to pick up a different one once in a while??? Not when it's always been sitting right next to me at the table, I guess.

I recently told the story of about 6 months or so ago I got whacked and let loose of my guitar almost spearing an o-scope screen. It really hurt, but it also fried my bridge pickup. I installed a new one probably a few months after. Must be I have some work to do in there, because its the cause of (most of) the ground swirl I hear.

This does not by any means fix my amp, but it certainly chilled it out a bit. I can still hear cruddiness there, but it sounds more to me like RF, and a tad of motorboat action.

Time to break out the scope and kill all unwanted upper kHz once and for all. I think that control on the cathode of V6 is contributing to some, if not all of the boating. Anyone seen anything like that little circuit jem? It's original to the BK Butler that the stage was adapted from. Any foreseeable improvements that won't overly affect its desired effect?

[jazbo8]

It's a bit hard to read the schematic, but better than nothing... V6 seems to have positive feedback phase cancellation, was that the intention? Or may be I got my eyes crossed trying to follow the wiring   Also, could you please summarize in one sentence what is the current status with the amp, this thread is a bit hard to follow with so many changes and suggestions made...

[Meat_And_Beer]

I also have each individual stage/ckt as its own sheet of paper that is much easier to read, this is the whole layout as is. If anyone want's a section up that's much clearer, just give me a shout.

The schematic does not reflect 22pf caps that now exist between the grid and plates of V1a, and V4a.

Problem summary;

Both channels have what seems to be rf/slight motorboating.

Channel 1 has a signal riding/coupling on it that makes the desired signal sound terrible. The 250k pot (p13) at the input of stage 3 / V6a can be turned all the way down, and you can hear just this unwanted hitchhiker. As you bring it up, the good signal comes in, but cluttered with the undesired one. (This issue exists from beginning to end of the preamp, it's just most obvious here.)

Channel 2 has its problem with the "bias" control at the cathode of  V6. (Lin taper) It's boating a bit. Its 2nd tone net off the plate of V6 has the treble control issue. It's the 500k with a 220k resistor across it, P15. All the way down, volume is good. Turn it up, treble increases, but volume decreases.


Sorry, a bit more than one sentence was needed... But that's about it in a nutshell. Other than that, its just happy tweaking!

Happy Fathers Day, gentlemen!

[jazbo8]

Sorry I just can't make out the parts that you mentioned, but it seems that the descriptions of Channel 1 & 2 were flipped in your latest post - just to confirm Channel 1 is the top row and Channel 2 the bottom row, right? May be you can highlight the section that you are having problem with, on the large schematic above. We'll help you get to the bottom of this eventually, but I suspect there are feedback loops from either mis-wiring or parasitic coupling given such complex build.

[Meat_And_Beer]

Hi Jazbo8,

I made a mistake, it's V6 that has the "bias control" on it's cathode with the motorboating problems. I said V5, sorry my mistake! Ever since I had to divide one 12ax7 into two 6av6's, the V's have thrown me off. (Cross-talk elimination) Everything's correct on the 'matics that I've provided here, I just slip up every once in a while.

Yes, Channel 1 is on top, channel 2 is on bottom.

I really don't think it's a wiring error, this amp was working. (I could, and always can be, wrong though) It indeed is a coupling that is happening somewhere. I suspected the fx loop switch, because it has some lengthy cabling attached to it, and large signals like to couple at those points. Also a concern for probability to me, was the "plexi mode" switch, for the same reasons. I've tried bypassing these with no luck so far...

I'll remind you I went through great lengths to electrostatically shield all cross-talking components. I had them playing together very nicely. Switches are my next suspicions.

I'll highlight and re-upload a new schematic and replace the above one. Attached here is JUST last stage of channel 2. Motorboating bias control, misbehaving treble control. (There were not an issue before, they're known good, but the bias ckt is known to be fussy with placement.)

[silverfox]

Would you be able to post a link to some recordings of what the amp sounded like prior to developing the problem? The circuit looks fascinating and I'd like to hear what came of the development effort.

I've considered using a tube mixer on, I believe it was a Plexi with two input tubes ultimately driving the power amp.

Glad to hear you have a decent Mentor. They're very hard to find nowadays... Fortunately the Internet just about makes them obsolete and where it doesn't, you have so many smart people it's possible to burn through them faster than they can indigestate, (coined phrase), on you. The collective knowledge of the machine is superior to machinations and you don't get the personality disorders associated with inflated egos. Doesn't sound like you've ever had to deal with that yet. But I digress; really interested in hearing the original amp as it may provide a clue.

I'd also be interested in hearing you play if. Your earlier post regarding signal processing for live performance says a lot. Here's a link to something I threw together for the guys here kinda quick so excuse the any rasp in the tune; clik the X to get rid of the kitty ad and you'll see the player.  http://mergatroid.minus.com/ldR28YJcsY3Z0

Fox.

[jazbo8]

Thanks for uploading the Channel 2 V6 picture, so what I thought was phase cancellation turned out to be just crossed wires when viewed on the large schematic... So this is a pretty standard gain stage - just curious why do you load both stages so heavily? 10k is a tough load, and the small slope resistor on the tonestack is better handled by the cathode follower IMO. The bias adjustment creates a pretty good size peak at LF when dialed down, perhaps that contributes to the motor-boating or instability on the low end? 2) did you check the treble pot to see if it is functional, it should not interact much with the volume control. And did you try bypass the whole fx section?

Jaz

[Meat_And_Beer]

Silverfox;

Right now my studio computer is 100% disassembled, we're moving at the end of the month. (I use ProTools for some asinine reason, it's an involved setup.) I only have two recordings of the amp when it was functional to speak of, really. The 1st one is when it was in the proto chassis, and I was using a parametric EQ to find exactly where a certain harmonic distortion/clipping was so I could smooth it out in the future.

The second is when I got the thing at it's best it's been in the real build. For some reason, I didn't save the session (All individual clips in a multitrack), I just saved the entire mixed down clip. It has about 3 or 4 harmonies and drums. This disappoints me because I was later trying to listen to JUST one track without all the others. I don't think I saved any of it, it was a quick demo for a friend down in RI, kind of just a "Check this out!" clip.

Both are of Channel 1, my favorite. I haven't had time to go through the entire amp as a play-through/demo yet, I planned on it after it's totally happy with itself.

I also have the feeling that the music I mostly play and listen to isn't very popular around here. (Tends to be hard, fast, and melodic)

You can check out a cover I did of Elton's Crocodile Rock here if you'd like:

Elton John Crocodile Rock cover... Metal punk style!

YouTube does a good job of swashing the life out of audio unfortunately.

I can provide a link to the project I had while living in RI with just a bassist. Some track titles aren't the cutest... I recorded, mixed, edited, all that stuff, all these songs. (There are no vocals, it's just music)

When my amp tech/teacher and I were first getting acquainted, he was also curious as to the music I played. I gave him a CD. And I quote, "Woah, way to knock us back into our wheelchairs!" That's the best indication I can supply. ;)

http://www.reverbnation.com/retox666/song/11698858-bar-fight-of-the-gods-pt-i



I really like the character of the rhythm guitar in that track you posted! The lead tone sounds very relaxing also. Nice work!

I can't wait to demo this amp, there is truly something for everyone in it.

[Meat_And_Beer]

Thanks for uploading the Channel 2 V6 picture, so what I thought was phase cancellation turned out to be just crossed wires when viewed on the large schematic...

No problem! Say the word, and I'll post a closeup of whatever is in question.

In regards to the slope resistor, The value choice was a mix between the original schematic and my tech's math. Should I try something more in the realm of 56k? It does have massive bass... pretty much everywhere. (It can always be toned down with the knob, of course) Reason being is this amp could double as a bass amp. It does go VERY low if you want it... I can even get my 2x12 with V30's to "woof". Normally this is a very mid/bright sounding cab with my other gear.

As to the heavy loading 10k's, this is getting into where I'm limited in my knowledge. I don't know. 

I re-situated the wiring pertaining to the bias control, it got me no joy. Should I adjust the 10k across the 22uf cap? I tried all sorts of different values for the cap, this is my favorite one, I'd rather not change that value if at all possible. But I'm up for anything that might sink this motoring boat.

Yes, the treble still squashes the output volume. I did try bypassing the FX loop, I jumped straight from the mixer tube out to the phase inverter and pulled the FX tube. No change. I still feel like there might be something I'm missing on that topic though. Something is evading me...

Little amp gremlin rascals... I'll get you!

[jazbo8]

Yes, I hate it when the amp worked before then just quits, then it is always something simple that botch things up, but finding it is usually not... Anyway, I don't think you should be changing any parts just yet, you got to find the root cause first. Your amp looks very well build with neat wiring and shielding, but it still does not hurt to chopstick around all the solder joints and move the wires around a bit to see if you can get the RF/motorboating to change pitch or amplitude, so at least you can isolate the trouble spots. Sorry couldn't be more help now...

[PRR]

What is this? It strongly resembles circuits which are notoriously unstable.

Since the amp worked before, it may not be an issue. OTOH it may be a lurking snake. And always possible it *wasn't* wired this way, one of your fixes "fixed" it, and it bit you.

[Meat_And_Beer]

PRR,

This is actually note for note the BK butler preamp, looks like we didn't actually change one component in there. Except, the bias cap, and additional 100pf across the plates. I actually own the actual unit as well. It's plates are getting something insane, like 13 or so volts. Mine gets +/- 60v at the socket.

This schem I'm providing is from Genz Benz, the now owner of Tube Works/BK Butler stuff. (It was emailed directly to me from them, not sourced from the net.) I had a different schem that was VERY similar, and the value of the cap on the similar one was the 22uf value that I preferred much more than the 4.7, and everything in between that I tried.

Notoriously unstable? Please, sir, do go on!

Jaz,

I've poked around with a skewer so much in this thing, it finally broke the other day! The motorboat got substantially more stable when I re-situated it's pot ground to the buss bar. Is there a possibility that when this gets really "fixed", that the treble pot will stop acting up? This has never been an issue for me, super strange. The treble, that is. I'm not really sure how I can go about addressing that. Parasitic oscillation...

What are some things that could be causing this?

-Lead dress/length
-solder joints
-what else?

So can it be said that we don't suspect any components have gone bad? Caps are 600v or more, and 1/2 watt resistors. Should be MORE than hearty enough I would think!

[Meat_And_Beer]

I got tired of thinking about it, and I pulled the BK out of the rack and started probing around in it.

Plates (In the actual, purchasable unit) are 11v.

(I've been very curious about this part) When "bias" is at zero, plate volts are 11. Turn bias all the way up, plate volts turn to about 2.5. This really freaked me out when I couldn't figure out why the supply in my amp was all over the place. Stupid knob! Glad to know its inherent in the circuit at least.

The components in the cathode bias network are indeed what is indicated on the actual BK schematic. Why then, when I auditioned caps for this, was I getting nowhere near the effect like it has in the actual pre? They (Bias knob) now behave identical... (Real BK Vs. mine, when my amp is having a good day.)

I'll set it aside while it's opened up for now. Any other things anyone would like to know about it? I'm not a whiz at PCB tracing, and it's dual layered too.

[jazbo8]

Now that we know where the inspiration for the circuit came from, some comments on its design - the Butler was designed as a stompbox (although a rack version is available), it is a starved-plate design and runs off +-12V, so when you use the same values as the original, it would not work the same way. Anyway I think the schematic contains an error, the 10k resistors should form a simple voltage divider. The way is drawn on your schematics now, there is no attenuation - not that it matters much since it should not be the cause of the RF and motor-boating. To make sure, you can always bypass the stage. BTW, have you pulled the tubes back from the power amp and find out where the RF/motor-boating starts?
                                                    

[Meat_And_Beer]

Now that we know where the inspiration for the circuit came from, some comments on its design - the Butler was designed as a stompbox (although a rack version is available), it is a starved-plate design and runs off +-12V, so when you use the same values as the original, it would not work the same way.

The unit in discussion is indeed the rack version, the RT-902. The (Partial) schematic I supplied is for the rack version.

The motorboating is on the bias control, I'm sure of it. Been there with the prototype. Channel 1 has no motorboating anywhere, but it has the same RF characteristics as does channel 1. I've pulled lots of tubes, searching...

Would it be helpful if I made a video for first hand action and explanation of exactly whats going on? Sort of like a hearing the words, and also being able to watch the lips move type of thing? I have just enough gear NOT yet packed up to be able to do that. I think I'm going to. I'll try to get to it very soon, within the next day or two.

[jazbo8]

The motorboating is on the bias control, I'm sure of it. Been there with the prototype. Channel 1 has no motorboating anywhere, but it has the same RF characteristics as does channel 1. I've pulled lots of tubes, searching...

I am sure we can get rid of the motor-boat problem by adjusting some of the parts in V6, so look on the bright side, now it is just a matter of tracking down the RF problem (very likely some HF oscillation), then you are set.

[Meat_And_Beer]

The bias adjustment creates a pretty good size peak at LF when dialed down, perhaps that contributes to the motor-boating or instability on the low end?

Do you have a recommendation as to where to start adjusting?

Would there be any adverse effects if I were to put a 22pf across the plate and grid of each preamp triode in there? Or do they need to be strategic? I haven't checked, but my instinct tells me that even if there is the 22pf in a stage, thereby not able to feed HF the next, it doesn't mean the next stage can't "hear" as high/fast just because the signal is being fed to it by something that is no longer able to.

Is that acceptable logic? I would think I want to 'limit' at least the first two or three stages in there.

[jazbo8]

First can you please verify if what I said was correct, can you control the amount of motor-boating when you turn the bias control? I only did a quick simulation, but it may not be what's causing the problem... Hate to harp on this issue, but if you pull out all the preamp tubes from both channels, leaving in just the mixer, PI and the power tubes, do you still hear the RF? If not, then you can put in one tube/stage at a time until the RF appears, then you know for sure where it started...

[Meat_And_Beer]

Jaz,

Yes, the bias knob was in control of the boating. However, since as previously mentioned, when I re-situated the pots ground, it helped immensely. Before, it would boat, after, I'd hear a "wubb" as I passed about 2 on the pot, if it were labeled 1-10 concentrically. Just now, I don't get the "wubb".

It is acting strange still. At zero and bringing it up, I can hear it increasing the hiss of the amp sharply, then quickly diminishing it before I even get to 12 o'clock.

I did reverse tube plucking as you asked, and it yielded the results I had assumed. V1, and V3 (first two stages of both channels) are the culprits for the dirty noise floor and RF. V1a, V3a, and V3b all have the 22pf grid to plate, and V3 is the worst. Even with having 1st and 2nd stages with this mod as opposed to just the 1st stage of the other.

V3 is channel 1
V1 is channel 2

In conclusion, the amp sounds pretty darn quiet given the settings I have it set at for this test, and with those two tubes pulled.

What is a better way to eliminate RF other than the grid to plate cap I'm doing? It's not all high freq stuff, there's a lot of low garble in there too. I think it's post three I made here in the beginning, the sound clip I attached a link to. It still sounds like that, but not as bad.

It goes to show I've been working on this amp a lot more than I should be paying attention to actually playing the guitar.

When we first determined the value of and implemented this 22pf grid/plate cap, we watched it on the scope. Adjusting the signal generator, at about 15k it would start to roll off, at about 20k the signal had shrank by at least 50%, and continued decreasing on the same curve as K further increased.

Thank you so much for continuing this battle with me, Jaz.

[Meat_And_Beer]

I'm adding a clear shot of stage(s) 1. Both are essentially the same, with some minor variants.

This is channel 2, as is. (Ignore the dotted capacitor on the 1 meg output pot)

For channel 1, subtract the 100pf across the plate resistor. Subtract the 22uf across the cathode resistor. The dotted cap on the 1 meg output pot is a 270pf.

Remember that the input jack goes to one single shared 68K, and then goes to both channel 1 and channel 2 grids of each "stage 1".

[jazbo8]

You are welcome, I see (sorry if I make you repeat yourself), so it appears that you may have too much gain and/or picking up interference at the first stage for both channels. Some quick things you can try, in no particular order:

- move the 1M grid leak resistor to before the 68k, i.e., it goes between the input pin and ground;
- remove the 22uF cathode bypass capacitors or substitute a smaller one say 1uF, that will knock the gain down for both stages;
- add a capacitor from the input grids to ground, say 100-270pF, this will roll off the HF.

[Meat_And_Beer]

Jaz,

I tacked in an additional 1m to ground, where the 68k, and two inputs for each channel merge. This didn't help with what we're looking for, but man, did it kill some hum! I notice no sonic difference with or without the resistor when playing a chord. (Just a quick test, but I like it! I think it'll stay there.)

I tried a 100pf across the 1m (effectively grid to gnd), no difference. I tried it on the socket to its circuits ground, nope.

Channel 2 still sounds better than 1 with respect to the RF/noise floor. Channel 1 doesn't even have a cathode bypass cap on stage 1, and it's the worst of the two channels.

I made up a little brass shield to go around the input jack and its resistors, giving it its own little grounded hut to live in. But that didn't help either.

Your suggestions were awesome, thanks!

[jazbo8]

Sorry to hear that you did not get very far, one more thing to check - if the amp is quiet without the guitar plugged in, i.e., both inputs are grounded, and the RF/noise only appears with the guitar plugged in, then I think you simply have too much gain, to verify this, you can bypass some of the stages before the signal hits the mixer to see if the amp quiets down beyond that I am kinda out of ideas...

[tubenit]

Reply 86

you can bypass some of the stages before the signal hits the mixer to see if the amp quiets down beyond that I am kinda out of ideas...

I think bypassing stages is a really good idea.  

Reply 21 was

Do you know how to safely jumper over sections?   IF so, try running V1a directly into the return of the FX.  Does that work?  Then add V1a & V1b into FX return. Does that work?  and so on Your goal by jumpering and bypassing sections is to locate which section creates the problem

With respect, Tubenit

[jazbo8]

Channel 2 still sounds better than 1 with respect to the RF/noise floor. Channel 1 doesn't even have a cathode bypass cap on stage 1, and it's the worst of the two channels.

This is a bit odd, since Channel 2 has more gain stages than Channel 1 - need to find out why. Also, with the mixer and/or channel volume control you should be able to mute one channel. Doing this should reduce the noise as well. so far, you failed to isolate the source of the noise, until you do that, you are still shooting in the dark... I know how irritating this could be, been there myself a few times, it sure takes all the joy out of amp building 

[Meat_And_Beer]

Jaz,

The amp sounds the same with or without a guitar plugged in.

This is a bit odd, since Channel 2 has more gain stages than Channel 1

They both have 5 gain stages, does a follower/buffer count as a gain stage? I wouldn't have thought that not knowing... But, not only that, stage 2 has cathode bypass caps throughout, channel 1 does not have a single one. A real nice comparison can be done on this amp in that both of their 1st stages are the same, (MOSTLY) aside from that cap. What I noted, is that a stage with a cathode cap has about 8 divisions higher gain than one without, seen with a scope. What is HEARD, is super different, also. And I, personally, am in love with that sound.

I am not opposed to lessening the gain on stage two, as it is a ravenous monster. But it's pretty cool. Stage one on the other hand, has just enough gain to my ears. I would weep if I had to tone that one down!

Tubenit,

I will definitely try the jumping. Why didn't I already?!?! I remember now, and it's just a bunch of excuses. I'm gonna dig into it. It may be a bit before I can give a solid progress report on this one... Empty boxes awaiting filling are my eminent destiny...

Gentlemen, I thank you very much for your support and efforts.

[tubenit]

Chris,

Just some thoughts for you.

I've done over 20 different builds that weren't clones and were somewhat quasi/pseudo kinda sorta "original" designs.  What that translates to actually is that I've simply repackaged ideas and innovation of other people meaning nothing I have done has been truly original. Even when I "came up with an original idea", ........ I learned later that someone else had done it.

I've tried ALOT of different stuff and spent 100's of hours experimenting.

There have been a couple of builds that I gave up on and cut my losses cause " I couldn't get there from here".    I scrapped those builds and used the trannies, pots and chassis to start over. I think Geezer (who is better at this then I'll ever be & is perhaps the king of innovation here) has gone thru that a couple of times. I view failure as part of the process to success.

Based on what I've been thru,  I'll offer four thoughts:

1)  I don't think you need 5 gain stages and I don't think a 5th gain stage is going to offer anything special. In fact, I think you will find
     it to have diminishing returns & create unwanted noise/hiss that is not useable.

2)  High gain amps need to have a very careful layout.  Without a good layout and grounding, you're going to have oscillation problems.
     You have a quasi point to point & layout board hybrid.

3)  You have stated repeatedly that the amp worked and sounded incredibly great prior to modding it some more.  IF that is truly the case,
     why in the world wouldn't you go back to that?  It is bewildering to me that you won't return to something that you viewed as true
     success?  

4)  IF you have created something that leaves these forum guys stumped in trying to help you, then I will argue it is too complicated to be
     truly be useful to you.  My guess is that you have a design flaw in the approach that you will not resolve without retracing your steps
     back to what previously worked.

My two cents.   With respect, Tubenit

It may be useful to you to look at this high gain amp thread to see what has been done on successful builds:
http://www.el34world.com/Forum/index.php?topic=15622.0

[jazbo8]

So both channels has 5-stages, I need to go back and look at the schematic more closely, perhaps I was thrown off by the long wire after the first stage in the top row, I thought you had skip a stage there, anyway, got to head out now, check in with you tomorrow.

[Meat_And_Beer]

perhaps I was thrown off by the long wire after the first stage in the top row

I thought that might have been it. Originally there was a weird tone switching circuit that my tech had devised for me. I didn't like it, so it went away leaving a big blank hole in the schematic.

Call me crazy, or thick headed, but I'm still determined. I'll totally admit to being a bit stubborn at times, too. But there is no way I'm turning my back on this build. It will get there, I just still have a lot to learn, and I'm cutting my teeth on an inadvisable first build.

I've got RF/ground issues with the fist stages, two post that are oscillating is some manner, and a coupled signal.

That's a short sentence. I'll get there! 

[Meat_And_Beer]

I jumped stage one from channel 2 straight to the PI. It's very clean, other than a slight overdriven/buzzy sound at all volumes on the pot.

(Link to sound clip provided below)

What is it? I'll tell you, and I'm very happy to have found this out. If I turn the volume down on the guitar a bit, it goes away. I use active pickups for the most part. When we drafted this, I guess that 1st 68k resistor it hits isn't enough.

I brought the volume on the guitar down to where the buzz went away and it was fully clean. I pulled the cavity cover off my guitar and read the pot. With respect to ground on the wiper, its 4.8k. Wiper to lug 1 is 17k (Which is the number I want, I think...) It's a 25k pot.

What does my 1st resistor now want to be? Is it additive? Whatever the common resistor is around the 85k area?

I could add an active/passive switch... Yay more switches!!! :) (Kidding)

So that rules out stage 1 channel 2. (Right?) I'll later check out its stage 2. I have a suspicion of a certain switch that's coming up real soon in the circuit...

Thanks, Tubenit, for really driving that suggestion home to me, and thanks for being part of my amp diary, fellas. :)

https://soundcloud.com/agormortisrr1/channel-2-stage-1-to-pi

[Ed_Chambley]

Chris,
Watching this happening.  Keep going.  Tubenit (Jeff) has some great information and it looks like you are beginning to narrow it down some.
If this is truly your first build, man it sure is industrious.  I too have built what I thought was a brand new idea only to find out that it was not new, but you will never get a new design without trying something new.

Get her running and let us hear it scream.

[Meat_And_Beer]

Ed,

Right on man!!!!! Thanks! :)

So a good update, I think I've localized it, For channel two at least. Weather or not channel 1 follows suit in response to the repair, will be seen. (Or, heard?)

From the guitar into stage two of channel two, and out of that stage two to the PI, sounded great. Then I fiddled a bit with what few knobs I had. I'm glad I still have commendable hearing, because what my ears picked up was a very high frequency. But that's all I noticed, none of the garbley mess that's associated with when I plug a guitar into the thing.

This stage does not yet have the 22pf plate to grid HF roll off cap.. So I looked more. There are probably 4 wires in this amp that carry signal that are not shielded at this point. Two of them were right there at this juncture... I combined stages 1 and 2 and sent them to the PI. The garble appeared, but distantly.

So I shielded the two wires, one about 1" long, and another about 1 3/4" long. (First two stages in an amp on steroids, not taking a chance any more.) and installed the HF cap.

Sounds better! But, it was still there. Remember me saying...

I have a suspicion of a certain switch that's coming up real soon in the circuit...
https://soundcloud.com/agormortisrr1/channel-2-stage-1-to-pi

The switch I was talking about was the "Plexi mode" switch. You can see it on the big schematic I provided, it sends stage two directly to the PI. With the Plexi switch off, I hear it, plexi on, it's gone! When switched on, the 1st stage volume can be about 11 or 12 o'clock and I wont hear it. As it's brought past that, it comes on like the dickens.

Which brings me to a question I've been wondering for a while, why hasn't anyone busted me on my switches not going to ground when not in use/bypassed? Bust my chops, people! I really am still a newbie!

It's either the plexi switch, or the fx loop switch that's coupling. Gotta be, right??? I have seriously tried SO MUCH stuff way before even posting here that I was beginning to rule out the switches/fx loop by default. Nope, I was right in accusing them. The whole loop of stuff got connected when the FX loop was finally incorporated. Bam, instant issue.

I'll get those to shunt what I don't want when de-selected, and see what's what.

3)  You have stated repeatedly that the amp worked and sounded incredibly great prior to modding it some more.  IF that is truly the case,
     why in the world wouldn't you go back to that?  It is bewildering to me that you won't return to something that you viewed as true
     success?  

Tubenit, I'd love to answer you on this. I mentioned somewhere in this thread that I have in fact, tried to reverse the FX loop finalization mod. I didn't have success, but I at the same time didn't want to go too nuts looking to do so anyway. (I must've been one try away or something, because I literally feel like I tried everything!) All of these effects/bonuses/tricks are in there to stay, and I knew there was a way to make them all work and be friends. In the back of my mind, I knew it was going to be this.

I needed a systematic way of smoking it out. And as simple as it sounds to most on here, the "Logical Blocks" that Glenjeff brought up long ago, then Tubenit and Jazbo8 reinforcing the zeroing in by way of stage by stage, finally clicked for me to not be so dumb.

This forum post has been my "diary", a much better one than I've been actually notating previously. Thanks for your patience folks, I did have a motive in my head.

I have a couple other ideas, but I don't want to diminish the gain that channel 2 stages 1&2 have because that will adversely affect the "plexi mode". Aside from that, it's past my bedtime! I had to get all of this out and onto the table.

Thanks everyone, good night.

[jazbo8]

The switch I was talking about was the "Plexi mode" switch. You can see it on the big schematic I provided, it sends stage two directly to ANOTHER switch (FX loop on/off) and then to the PI. With the Plexi switch off, I hear it, plexi on, it's gone! When switched on, the 1st stage volume can be about 11 or 12 o'clock and I wont hear it. As it's brought past that, it comes on like the dickens.

It seems like you are finally getting somewhere... from the big schematic I can make out a DPDT switch, that bypasses everything, and sends the output from the second gain stage of Channel 2 directly to the PI, so is this what you mean by Plexi OFF? If so, the problem is kinda odd, since you have a lot less gain stages adding noise in this mode.

As for grounding the switches, I see with the Plexi switch OFF (if I read it correctly), the input of stage 3 in Channel 2 isn't grounded, but since the switch bypasses everything after that, it should not matter, of course, it would not hurt to ground it.

Also you are saying, with the Plexi switch ON and turning Channel 1's volume pass 11 O'clock, the noise appears, what about Channel 2, does turning it up to 11 O'clock, does the same thing?

[Meat_And_Beer]

Plexi mode switched on sends the end of stage 2 straight to the PI. Switched off, it goes through the rest of the circuit normally.

Okay, so now I'm doubting my previous post. Dodgy amp gremlin crap!

Channel 2, stage 1 only, straight to the PI. I'm listening to it now, it's faint, but the instability is there, and intermittent. When I bring it's only volume pot up, hiss it gets louder, then softer at the very tail. When it gests softer, is where the gross unstable fluctuating sound chimes in. This first stage must've been having a good night last night or something, because I didn't hear it then. 

Channel 2, stage 2 only, I just recorded and scrutinized. I killed 60 Hz, and sloped some low end so I could discriminately listen to whats going on in the rest. I really can't hear in there what I can in stage 1. It seems to be ok...

I thought it was the switch, maybe because I wanted to, but stage 1 is the real BS'er here.

I have not checked out channel 1 yet, I was hoping to wrap my head around this channel first, then apply what I've found to the other.

...Also, with the mixer and/or channel volume control you should be able to mute one channel.

Yup, and double yup. I can silence either channel with its final volume, or I can shunt it all together with the channel select rotary knob.

Crunch time for moving... Can't.... put the soldering iron... dddooowwwnwnnnnnnnnnn

[Tone Junkie]

Meat_and_Beer you can take your first bypass cap down to 2.2uf instead of 22uf it will help with some of your ocsillation problems it also will tighten it up for any palm muting stuff you might do.
You have a 1k there, At 820 R and 2.2uf your right at 80hrtz which is the bottem E string in standard tuning with 1k and 2.2uf you should be down around a drop D tuning, I could do the math and figure out exsactly were,but thats close enough .
I know engle and a few others use 22uf but they keep a balance in othere area,s . At 22uf  you are taking it into the 1 to 2 hrtz range which is lower that your transformer can play . It cause.s instability issues in high gain amps where in lo gain amp's like a classic fender it just makes it sound fuller, even then i never go above 4.7 uf, except when building classic circuits. Im talking strickly the first stage you hit.
 You could also go 2.2uf with a 4.7uf on a switch to add extra for bright sounding guitars.
You are wrong about thinking some of these guys wouldnt understand your music I havnt been in a metal band since the early 90,s and yes that makes me old. But there are a few guys here who dable with a high gain build once in a while. We play mostly classic rock now because thats what all are friends play (LOL).
Bill

[Meat_And_Beer]

Bill,

Thanks so much for that informative response! Most of my guitars are in drop C (CGCFAD), one in standard 440, and another in B standard.

I was just hinting/warning that the stuff I play might not be a lot of peoples cup of tea. I was hoping that I was at least semi-wrong about that one. ;) I'll probably soften up as the years go by two, seems to be the trend!

I'll give it a shot today! Lots of stuff going on, but it'll be quick, and I'll report back. Thanks again!

Chris

[Meat_And_Beer]

The closest thing I had was a 4.7uf, so I tacked it in there. No change. (I listened to unmodified just prior)

No bypass cap, no change either...

Man, there is so much I have to learn! The ways to tailor your circuit to your playing/tuning. Even just to get something to end up at precisely the necessary voltage for things...

Jeepers tube amp books are pricey. I feel like I'm in college again!

What this sounds like can be heard here:

https://soundcloud.com/agormortisrr1/horrible-noises/s-oI33f

HOWEVER, it has gotten a bit better. (A bit!) This clip is from the entire channel (Either of the two channels, and it makes no difference weather a guitar is plugged in, or the jack self grounds with no input)

What I'm hearing at the output of just the first stage is nowhere near as nasty in compared to what it sounds like throughout the whole circuit, like the link above. This makes me think two things;

1) Stage 1 is where it is originating from. Given that both channels are ALMOST the same, possibly it's something inherent in design.
(But it never gave me this much crap before... Or much of any crap at all!)

2) It's mechanical. A cold solder/bad ground connection somewhere.


Thoughts? Pointing fingers?

There is definitely RF in there. (But does weak ground cause RF? Does a cold solder, depending on where it is, cause RF?)

With more and more of my test gear getting packed/relocated, I can't tell you exactly what's going on precisely in respect to what frequencies are being allowed in there, and where. I WILL be able to in about a week or two though! To the extent of my ability, that is. Which doesn't TOTALLY suck, at least.... Well, a little.

Thanks for the beautiful support, team!