Updated my schematic finally !

[daz]

I finally edited it to reflect all the latest changes and would like to see what Chuck Steve and you others that have helped me out think about some of the sorta unorthodox things i've done that seem to have gotten me some really articulate yet springy tone that has me playing this thing a lot. one of those "can't put it down" things. as you can see i no longer hane the cathode caps on the PA switchable to take them out of circuit because what i did there gave me the tone i liked with them out of circuit w/o having to do that. I just put a 33uf on one side and left the 22uf on the other. The imbalance really did wonders. I also put the variable NFB pot back in, tho if i decide i like one setting all the time i'kk put a set resistor in. And as chuck mentioned i thought i'd try more imbalance on the PI too, so i added a 120k in place of the 100 so one plate has a 120k and one a 82k.

I'd like to know what you all think about those or any other design aspects you have thoughts about you feel i might benefit from. But i'd especially like to know what you guys think about the imbalanced PA and PI and whether you think i should do anything different or different values, etc. Also, i will likely have to try chucks idea about how to wire the presence again because right now it's all in the last little bit of the pot's throw. But a smaller value pot doesn't take the presence fully out. Si chuck, i gotta try that again because it's just too hard to tweak the presence like this.
http://i24.photobucket.com/albums/c3.../myEL34new.jpg

EDIT: forgot....steve, if you read this can you tell me what you think i could do to rid the amp of the motorboating i mentioned? However, realize that it doesn't do it while the amp is on, only right when i turn it off or to standby. then it will give a few putts as it goes off. Guess it doesn't really matter since it doesn't cause any problems while playing the amp, but it bothers me.

[Chuck H]

Very nice. No BS, just like your design. The imbalance on the cathode bypass caps could catch on. Did you try the PI imbalance on a switch for an on the fly A/B test?

You said the amp motorboats briefly when put into standby or shut off. When you switch your amp off do you flip the standby switch at the same time? Or does it motorboat when you just flip the power switch?

Do you have bleeder resistors across the main filter in this amp?

The only thing I would mention is that the cap in the presence circuit is intended to block DC from the adjustment pot. Ideally it would be on the input side of the pot. But if you had problems with it on that side leave it alone. Enzo made a good point in another thread and mentioned that as long as the cap is in the circuit no DC can flow, so what's the difference, right?

Chuck

[daz]

You said the amp motorboats briefly when put into standby or shut off. When you switch your amp off do you flip the standby switch at the same time? Or does it motorboat when you just flip the power switch?

It does it no matter what...standby, power, or both at the same time. I think this may have started around the time i re-arranged the PSU at the first can abd choke. Before i had it with the OT center at the first side of the 23/23 can with the choke also there, then the other side of the choke going to the other side of the can. Now i have the can by itself paralleled together and the OT center ot the output side of the can and the choke there to and then onto the screens. It didn't start right after i did that, only later. But i suspect it set things up so that any little thing could set it off and then did. But it sounds better with the PSU like that so i'm leaving it unless it gets worse. From what steve said and from the fact that it's been a lot worse and even gone away then return without making any changes suggests to me it's right on the verge of not being a problem and one small change somewhere like changing a resisitor by a few ohms may stop it.

Do you have bleeder resistors across the main filter in this amp?

No, what do those do and how do you wire them?

The only thing I would mention is that the cap in the presence circuit is intended to block DC from the adjustment pot. Ideally it would be on the input side of the pot. But if you had problems with it on that side leave it alone. Enzo made a good point in another thread and mentioned that as long as the cap is in the circuit no DC can flow, so what's the difference, right?

I was going to change it today to the way you showed me with the cap going from wiper to ones side and that side to the mail NFB point, then the remaining side to ground. When i did that before i recall the range was more spread out, so i want to do it just for that reason.

[daz]

Oh, by the way Chuck, did you at one point say that if i wire it like you showed in that jpg that i could use the 5k pot and it would be able to take the cap out at full ccw?

[Chuck H]

Yes, at fullCCW the cap is out of the circuit. But it's important to note that it must be wired just like the image I drew. There would be no more 4.7kR shunt resistor. The 5k pot acts as the shunt resistor. So you would actually be changing the value of your shunt resistance by 300 ohms. It shouldn't even be audible.

Bleeder resistors allow the amp to drain the caps when the amp is shut off. They are always a good idea. Especially if you work on it alot. Safety and all that. The standard arrangement is two 100k to 220k 1 watt resistors in series across the + and - leads of the first filter cap. On my last build I just used a single 220k 3 watt resistor and it works fine. This could help with your motor boating problem too. It seems that something related to the power supply is causing it. Which is confusing to me. I wonder if something isn't wired cattywhompus in there to cause this. Why would voltage rapidly dropping from all the tube plates cause motor boating? Anyway, adding the bleeder resistor/s may change the phase angle a little and also change the voltage drop rate. Either might help.

AAaahh. I just took another look at your schem and now I get what you were saying about how the PSU is wired. You should have another filter between the downstream side of the choke and the 2.2k resistor. I'd bet a dollar to a donut that would end the motor boating. The only rub is that it will alter the tone of the amp. Probably for the better but that's subjective of course. It can't hurt to try it out. Since the amp is working now and you like the tone this cap doesn't need to be large. 22uf would be fine. The Fender Prosonic uses a 4.7uf cap for this filter. The smaller the cap the less smoothing of the screen supply you will get and less effect on the tone the way it is now. So if you find a 22uf cap stops the motor boating but you don't like the change in tone you could try a smaller cap. Many amps use a largish value here. A well regulated screen supply will give the amp more power and a tighter sound. But in a good way because a poorly regulated screen supply can actually sound garbled and warbley.

Chuck

[daz]

Thanks Chuck. All good info. I will try the cap there, tho i'll have to pull from my other amp because i don't have any other high voltage caps. But i'll give it a go and even if it doesn't help the boating maybe the tone will improve even more. I'll let you know what happens. Thanks

EDIT: no go. It didn't stop the MB or even change the tone as far as i can tell. I used a 20uf cap from the screen side of the choke to ground.

[loudthud]

So there is no filter cap at the screen node?

[daz]

No, but i just put one there and there was no change so i'm not sure why i should. If i thought i should i would. But i had to scavenge one from another amp to try it and since there was no change in any way i would rather not order a cap which with shipping will cost me probably $20. If i find a difference i will, but i couldn't hear any diff and the MB didn't stop. I will try it later on when no one is home and i can crank it to see if i notice anything when loud.

[tubeswell]

Originally Posted by daz

No, but i just put one there and there was no change so i'm not sure why i should.

The output filter cap (or screen-node) is there to smooth out the B+ bumps leftover after the reservoir cap and choke/resistor have done their jobs.

Originally Posted by daz

But i had to scavenge one from another amp to try it and since there was no change in any way i would rather not order a cap which with shipping will cost me probably $20. If i find a difference i will, but i couldn't hear any diff and the MB didn't stop. I will try it later on when no one is home and i can crank it to see if i notice anything when loud.

Was the one you scavenged from the other amp in working order? Motorboating is usually caused by bad filtering. On the face of it, what you have in your schematic is the PI plates interacting with the power tube screens, because there is no filter cap between the 2k2 supply resistor and the choke. So all the voltage swings being exerted by the PI plates are being reflected back through the 2k2 resistor and into the screen node (at least AFAICT)

[daz]

OK, well you guys know what you''re talking about so i'll put one there permanently. It didn't seem to do anything so worse case scenario: peace of mind.

Ok, i updated the schematic to show the new cap. You may have to refresh when viewing it or it may show the old one still in your computer's cache.

[loudthud]

I'm surprised the absence/presence of the screen cap didn't make an audible difference. Does the amp still motorboat with the feedback disconnected or at any position of the feedback pot? I noticed you removed the switch to disconnect the feedback. Low frequency phase shifts can sometimes make feedback unstable.

[daz]

The motorboating stopped, but not because of the added cap. It stopped after i rewired the presence pot as chuck described. But it never did that before with the pot wired as it was, so that has me stumped. I guess some other change cause the presence wiring to help cause it. I removed the NFB defeat because i never really used it. It was much too bright/hard and in your face like glass.

[daz]

It's baaaaaack ! yeah, for no reason the motorboating at shutoff is back again. Screw it, i'm just gonna leave it. I also noted that with the cap I added the amp doesn't sound as good. It's brasher, brighter, lost some of the transparency in the high end, which is what i was really digging. It was plenty bright yet not thin or harsh. So i'm just gonna remove it. I just don't see what i should leave it is the amp sounds better w/o it unless it's gonna blow it's top.

EDIT: look, i know i'm an idiot ok.....let me make that perfectly clear as ol' Dick Nixon once said. My schematic is wrong ! I in fact DO have a cap there already ! It's been there since i first built this thing, and all this time i never noticed the mistake on the schematic. See the 22uf right after the PI? Well, in the amp it's actually where i put the 47uf when i added that cap today. No wonder the amp sounded fine and in fact better when i removed that cap i put in today ! Well $hit.....i never said i knew WTF i was doing, did I?

[tubeswell]

So can you please describe again when you get the motorboating (only at shutoff?)?

[daz]

Yes. When i turn it off either with standby or power switch. At that point it starts and dies off in a second as the amp powers down. It doesn't do it at all while i'm playing.

Chuck....i think i need to go to a 10k presence because the 5k doesn't shut it off completely. I will have to check that 5k to see if it's linear and if so get a audio because the range is still not distributed well over the length of the pot's throw.

Also, i updated the schematic again and now it should be 100% accurate, as i fixed a number of things to reflect exactly how it is now. Refresh your browser when looking at it or you may still see the old one via your history/cache, etc.

[tubeswell]

Well that sound sounds just like your filter caps discharging and it is normal - nothing to worry about. A few of my amps make a similar sound on discharging

[daz]

but the thing is, it is the remnant of what WAS a problem at one point. At one point i tried a different value of the PI input cap and it motorboated constantly with the amp on. After i put the original back it only did it at shutdown. So whatever is causing it is on the verge of doing it and i suppose at shutdown it's just a but more vulnerable to it as the caps discharge. Doesn't matter i suppose.

[pdf64]

Re the presence pot, another option would be to use a 50k ohm inverse log pot intended for a fender tremolo intensity control - the bug opto coupler type. Should be near out of circuit on '0' but with good resolution as it's turned up - 50% should be 5k, 75% should be 500 ohms. And easy to source.
Re motorboating, have you tried different 12AX7 tubes? Sometimes an odd one can push normally stable amps close the the edge. Peter.

[daz]

I'll keep that in mind, tho i think what i'll likely do is a 10k pot with the trace cut at one side so it's fully open turned all the way CCW. As for tubes, i don't think thats it but i'll give it a shot. thanks

[Chuck H]

Daz,

The way you have your presence wired is NOT how I drew it or described it. right now you have the pot AND the original shunt resistor acting in parallel as the NFB shunt. This substantially decreases the amount of NFB from where you had it. This could also be why you didn't get linear operation from the circuit originally. Look at the drawing I have attached. The idea is that the pot serves double duty as the shunt resistor AND the presence adjustment. This will return your feedback ratio to where you had it the last time you said it sounded good.

Chuck

P.S. I'll be out of town for a few days.

[daz]

Ahhh....i see, and i did it. Works great, thanks Chuck. The MB stopped today but i'm sure it will be back. With all the changes i made lately i now have the amp at work where i can turn it up and i'm finding the truth about these tweaks. Some are definatly good, others i may be changing. One thing thats been bothering me for a week is i think i may have a bad OT ! remember i posted about the other amp like this one my friend now owns and the OT blew? Well, just a couple days after that i may have ruined mine by starting it w/o speaker. i played a chord and realized the speaker was unplugged and turned it off immediatly. Is it possible to damage the OT that quickly? I didn't think so because AFAIK it has to be on with signal for long enough for the windings to heat to the point the windings melt, right? yet my amp's volume seemed much lower after this happened. so when the new OT for his amp arrives i will tack it into mine to see. i may be ordering may second replacement OT within a couple weeks !

[tubeswell]

Originally Posted by daz

One thing thats been bothering me for a week is i think i may have a bad OT ! remember i posted about the other amp like this one my friend now owns and the OT blew? Well, just a couple days after that i may have ruined mine by starting it w/o speaker. i played a chord and realized the speaker was unplugged and turned it off immediatly. Is it possible to damage the OT that quickly?

Nah 1 chord won't do it. I've switched on amps that I've modded in the past and started to play before I realised the speakers weren't hooked up, and the OTs didn't get wrecked. The energy/heat has to build up for a little while before the current can start to interact with the insulation

[daz]

Thats what i thought, but it really seemed a lot quieter after that. I have to clip the new one in to be sure because this thing was too loud to turn way up w/o killing my ears. Now it's easy to handle at full volume. And this happened b4 i began messing with the PA and NFB so it's not a gain change from tweaking those things.

[lowell]

Very nice schem. May I ask what you use to draw 'em?

[Enzo]

RUnning a tube amp with no speaker doesn;t ruin the transformer with heat. it can happen in an instant.

What happens is the voltage spikes that are reflected back from the unterminated secondary can arc your transformer windings. or potentialy the tubes. it all depends upon how large a spike is generated, and whether or not it punctured the insulating layers. It might happen the first strum, or you might get away with it for a long time. But nothing has to build up.

The longer you play that way, the more opportunities the wavefrom has to make peaks, and all it takes is one that crosses the magic threshold inside there.

[tubeswell]

Originally Posted by Enzo

RUnning a tube amp with no speaker doesn;t ruin the transformer with heat. it can happen in an instant.

What happens is the voltage spikes that are reflected back from the unterminated secondary can arc your transformer windings. or potentialy the tubes. it all depends upon how large a spike is generated, and whether or not it punctured the insulating layers. It might happen the first strum, or you might get away with it for a long time. But nothing has to build up.

The longer you play that way, the more opportunities the wavefrom has to make peaks, and all it takes is one that crosses the magic threshold inside there.

Oh well - seeming as how I am inviting trouble, can I change my story then your honour? I can see how if you have a big enough voltage spike it will puncture the insulation, but I was generally thinking of spikes that aren't that big, but are made more destructive with the addition of heat. Does that sound any better? (Not that I'm expecting to be let off the hook .)

[Enzo]

I'll have to think about it. I'm closing up for the night and going home.

[daz]

Very nice schem. May I ask what you use to draw 'em?

Photoshop. What i do is make the symbols like resistor, cap, etc, then after i've made one the rest i can copy and paste. You can select anything then copy and put it wherever you want, rotate them, etc. With certain functions like that you'd be surprised, but it can actually be easier than many dedicated schematic apps, at least the few i tried. once you figure out all the things you need to move things around any way you want it's quick and easy.

on a side note, that schematic no longer represents it's current circuit. It has changed a lot since then. at the moment it's probably better than it's ever been and i've been mostly just playing instead of tweaking lately with few exceptions. When it sounds this good it's hard to stop. (much thanks to Chuck who really sent me on the right path not long ago)

[lowell]

right on, that seems like a great way to do it. only drawback is not being able to test the circuit and view voltage/current characteristics in the software before even building it. However, so far I find building a circuit to be the most reliable way of deciding whether or not it sounds good and works.