A Million Questions - Traynor mod/rebuild

[gazinkus]

Gentlemen:
I have just acquired (for relatively little) an old Traynor Voice Master YVM-1. It's a 4-input dual-el34 PA head that has already lost its "vintage" value. I intend to modify further. I'd like to end up with a very versatile channel-switching amp with a number of classic tones, and I have some design ideas. Further background: I'm an EE - but one mostly skilled in systems and digital stuff, and a multi-instrumentalist with years of playing experience. I'm a near-rookie in analog/tube circuitry design, but a decent troubleshooter. I play a range of stuff from country to medium rock, so I tend to prefer "classic rock", clean, and bluesy tones over mad distortion. My Fender Blues Junior (stock) is great, but not big enough for all my applications.
Here are some of my questions:
1) The head has 3 12ax7's in the preamp and 1 in the phase inverter section. All of my design ideas eat 12ax7's. So, I'm trying to free up as many as possible. Anyone ever try changing out the phase inverter 12ax7 with an op amp circuit? Given that the input and outputs of this particular phase inverter circuit are AC coupled, if I found the appropriate op amp circuit it seems I could free up a 12ax7 here. What is the gain of a typical phase inverter circuit? What impact would this have on overall tone, of say, a standard Marshall-type design? I've never heard of phase inverters being key to tone unlike, say, power amp tubes, preamp design, speakers, cabinet design, or OT's. So, if an op amp circuit had the headroom....
2) Assuming that the above is a bad idea for some reason, I sure would like to add a couple 12ax7's to the chassis. My understanding is that Traynor overdesigned the trannys, so I'm hoping that the power supply will have the extra juice. I know I have to deal with the max plate currents and the heater currents. Anything else? This guy has a Hammond 78633 - and I haven't found specs for it. Anyone know what it'll put out?
3) Is there any benefit/detriment to going DC on the filaments? Fuchs claims to do this. It seems to me that it would be an easy mod to lose some hum in an amp.

Any help will be appreciated, and will probably be followed by more questions. I'm in full-blown learn mode here.

Thanks,
Dennis

[Enzo]

I wouldn't worry about the gain, I am sure an op amp can be made to have more than enough, but what will you do if the output tubes want to see say 30v of signal?

Persobally I would rather make a hole and mount another tube than use the op amp. Aside from any level issues, there is also the matter of making an entirely separate power supply for it and then regulating that. Believe me I am no tube snob, but I don't like the idea.

You might think about why your designs use so many tubes. Try having tubes do double duty. For example a crunch channel and an extreme channel can be the same channel with some gain structure elements switching in and out. Still want separate volume controls for them? FIne, have a little relay select controls. Now two channels use the same signal path through the amp.

Another trick Peavey and others use is to have a basic clean channel, then after the first stage or two, switch in a branch path through a couple more stages of a 12AX7 and then return. Same basic signal path, but kick in the dirt channel and a whole extra tube is there with its two triodes. When they do that they also relay switch control,s and some other tone shaping and gain controlling stuff.

Three 12AX7s is 6 triodes. That is a lot to work with.

In some amps, the phase splitter is the furst stage to clip. PV C30 for example.

12AX7s wired 6v draw .3 amp a couple extra tubes adds a half an amp. I wouldn't sweat that. The B+ draw is minimal. Those triodes usually conduct a milliamp or two, that's all. SO two more 12AX7s means maybe 6ma extra off the B+.

DC heaters is one way to reduce hum, but for me the question is how much hum do I have to start with. I wouldn't want to waste my time applying a solution to a problem I don't have. Keeping your heater wires twisted, away from signal lines, and grounded goes a real long way towards hum reduction. Ground teh heater wires by grounding the center tap of the winding, or making a faux center tap with a couple 100 ohm resistors.

You can always go back in and make it DC later. Generally only the ifirst stage or two needs it anyway.

[Steve Conner]

If you're looking for ideas to save tube sockets, see RG Keen's MOSFET Follies series of articles. To sum up: Op amps no way, discrete MOSFETs maybe.

But as an EE and programmer myself, I should warn you that the design techniques you learnt for FPGAs and cruise missiles don't necessarily translate to tube musical instrument amps. The most desirable tube amps ever were designed by trial and error by a radio repairman and a drummer. They had no option but to be inexperienced, because rock guitar tones didn't yet exist, and they were in the process of inventing them.

In short, why not forget the million questions and just get tinkering...

[gazinkus]

I wouldn't worry about the gain, I am sure an op amp can be made to have more than enough, but what will you do if the output tubes want to see say 30v of signal?

Aye, there's the rub. When I asked the question, I was about 6 hours earlier in my research. It appears that I have to drive to at least 37V of output swing. That would be hard to come by in an op amp.

You might think about why your designs use so many tubes. Try having tubes do double duty...

Yeah, I will be doing that. Right now I'm considering a "Dumble" set of channels and a "Marshall" set of channels, each with the possibility of footswitching in an overdrive structure. The Dumble Overdrive setup requires 2 12ax7s - one for the clean side and one for the dirty side. A Plexi-type Marshall requires 1.5 12ax7's, plus one for a switchable gain stage. So that's 4 total. The big problem for me is that I want to keep the tone stacks separate - for two reasons - to have better user control and to maintain the voicings. That leads to issues in itself, e.g. real estate, which I'll have to solve by ganging pots. But who knows? Once I get to tinkering, I might decide that the Dumble side is all I ever want...

In some amps, the phase splitter is the furst stage to clip. PV C30 for example.

Another of my research finds last night was an article from a fellow at Groove Tubes. His contention was that MOST of the power amp character in a tube guitar amp comes from the phase splitter. If that is the case - or if even any appreciable amount - I sure don't want to go solid state there. Even if it sounds "good", it won't sound like what I want.

Yet another find: It appears that my year YVM-1 has a totally different phase splitter than the original (for which I had a schematic). I found a schematic for the later models. (I don't have the amp yet to know for sure, but it sure seems the later schemo is a better fit.) The original amps had a cathodyne structure preceded by a buffer stage. It took me a while to figure that out from the schematic, although it was clear that it WASN'T a long-tail design. But, the latter amps, e.g. mine, use a pretty standard long-tail.

12AX7s wired 6v draw .3 amp a couple extra tubes adds a half an amp. I wouldn't sweat that. The B+ draw is minimal. Those triodes usually conduct a milliamp or two, that's all. SO two more 12AX7s means maybe 6ma extra off the B+.

That's what I figured. Also, the "new" (1969) YVM-1's had a different PT than the older ones. This one appears to have a Hammond A1310 or 81384 (I still can't find specs). The main difference appears to be a separate secondary for the power tube bias circuit. But I can't see where a couple more 12ax7's constitutes much of a load on either version. Being a rookie at this, I sure can be wrong, so it's nice to have another opinion.

DC heaters is one way to reduce hum, but for me the question is how much hum do I have to start with. I wouldn't want to waste my time applying a solution to a problem I don't have. Keeping your heater wires twisted, away from signal lines, and grounded goes a real long way towards hum reduction. Ground teh heater wires by grounding the center tap of the winding, or making a faux center tap with a couple 100 ohm resistors.

You can always go back in and make it DC later. Generally only the ifirst stage or two needs it anyway.

Good advice. If it ain't broke, don't fix it...

Many thanks,
Dennis

[gazinkus]

If you're looking for ideas to save tube sockets, see RG Keen's MOSFET Follies series of articles. To sum up: Op amps no way, discrete MOSFETs maybe.

Did that last night. I came to the same conclusion. But, after reading elsewhere that the phase splitter contributes much to the character of the sound of the power amp section, I punted on the whole idea.

But as an EE and programmer myself, I should warn you that the design techniques you learnt for FPGAs and cruise missiles don't necessarily translate to tube musical instrument amps. The most desirable tube amps ever were designed by trial and error by a radio repairman and a drummer. They had no option but to be inexperienced, because rock guitar tones didn't yet exist, and they were in the process of inventing them.

Yup. I used to refer to analog work as "FM"... Freakin' Magic. Loved the sound; hated the design aspects. Digital was so much easier and more interesting to me. I've come over to the Dark Side... And, on top of what you say, much of the desireability of tube amps comes from their flaws in a design sense - the facts that they didn't quite do what they were intended to do because of real world limitations. Distort? What engineer in his right mind would design an audio amp to distort? Compress? Ditto.

In short, why not forget the million questions and just get tinkering...

Well, part of the reason as that I AM an engineer by training. It's my nature to research and plan and all that. And part of the reason is that the last time I tried something like this - I had an old tweed Bassman head (probably a chop) - that I wanted to make into a Boogie. Stupid idea. I got as far as ripping out the guts before I got diverted. This was in the late 70's when a tweed Bassman was hardly the Holy Grail it is today [I paid $50 for it...]. It eventually was thrown away as useless garbage (!). I'd like to not repeat that misadventure. And, lastly, I have a million other projects to finish, so I'd like to not make this my life's work. I'm still a gigging player, so I want a tool I can use.

Thanks again,
Dennis

[Steve Conner]

I'd like to not make this my life's work. I'm still a gigging player, so I want a tool I can use.

Well, I contend that tinkering by trial and error will get you an amp that fits your playing style quicker and easier, and more enjoyably, than trying to work it all out beforehand.

You can even go rehearsing and gigging with it before it's finished. I like this approach because I get to try my creation in the context of a band, and on a typical night with several acts, I get to hear it with other amps, and also other musicians will play my amps and give me feedback. I'm always afraid they will pop a fuse or electrocute someone and get me sued, but so far, fine.

I had an old Selmer T'n'B 50 head that I used like this for years, modding as I went, until it was transformed into a totally different channel switching amp with spring reverb. It wasn't until the original resistors and caps started to fail that I went for the complete gut and rebuild approach. Of course I managed to get it apart fine, but I've been struggling to find the time to put it back together again! Hopefully it won't go the way of your Boogie Bassman

steve

[gazinkus]

Ok, boys...
The amp arrived today, and I began the process of tearing it down, cleaning it up, seeing what I have. Naturally, I plugged it in and tried it, as is. Here's the results:
1) This amp has been modified in the past by an amateur (i.e. worse than me, and I'm a newbie to these things). Sloppy work, poor materials. A lot of that is history, since I'm planning on a total rework anyway.
2) I plugged it in, pretty much fresh from the mailman. The tubes were shipped in it. It fired right up and played. Not well, mind you. Lots of hum. After about 5 minutes, there was a couple pops, and the 4A fuse blew (it is rated for 3A). Ok, I said. I figured something like that. Let's tear it down. And when I did, I found two sheetrock-type screws floating around inside the case, a broken 12AX7 in the first tube slot (unused in the current configuration, but surely hooked to the master volume mixer circuit), a 3-wire power cable in which the ground wire had been lifted from the chassis and was dangling around in the breeze, apparently original Mallory electrolytics (the two-in-a-can type), apparently original 6CA7's (RCA), apparently mostly original 12AX7A's (Phillips, date codes from '67 and '80), no obvious burn marks or fried parts anywhere, no more loose wires, some really ugly (and hum-grabbing) signal cable lead dress, madcap grounding, but nothing terribly odd to report.
3) To paraphrase Jimmy Buffett, being charitable and cautious, I decided to give it another go. This time, the chassis is sitting out on my work table, I have a 3A slo blow fuse in it, and a good 12AU7 sitting in the V1 slot (which isn't being used; I suppose that I could have left the slot empty to the same effect, but I had the 12AU7 sitting here). As per the prior trial, I was plugged into a speaker cabinet, and my Frankenstrat was plugged into channel 3, one of the 2 routed to the front panel in the place of the original channel 1 and 2 pots. (Channels 1 and 2 are dead, so V1 gets no drive signal from anywhere; the 2 remaining input jacks go to channels 3 and 4, which drive V2, a 12AX7a). This time, I fired up everything on standby and observed for about 5 minutes. No problems. No smoke. Orange glow from all the filaments. Flip on the power switch (no input, channel volume 0, master volume 0). Nothing to report but a little hum. Turn up the master volume to 5. More hum. Turn up the channel volume to 5. More hum. Pots are dirty, so there's some scratchiness. Turn up the treble... more hum and some crackle. OK, plug in the Strat. Plays OK for a while, maybe 2 or 3 minutes - pretty thin, but I wasn't expecting much from the cab, nor was I pushing anything. Every once in a while I would get a pop and a bit of crackle. Still lots of hum (I use noiseless pickups). Then, all of a sudden, one of the 6CA7's glowed medium-brightly (orange, like maybe the filament voltage went to 12 for about half a second) and the fuse blew.

I figure the following: First, the B+ electrolytics need changing. Really bad. Second, it appears that the trannys are working fine. Third, the tubes are ALL old and will need replacement. Fourth, the signal feed wires in at least the early stages need cleaning up, as do the grounds everywhere. But none of these really explain the time-delayed fuse popping. In my digital experience, time-delayed problems are related to heat, and heat in tube amps is mostly in and from the tubes - but that may not be true here in the analog world.

Oh, another data point. When the fuse popped, I pretty quickly turned the chassis over to look for singed parts (after unplugging the power cord). Next task was to discharge the electrolytics. If I was going to be digging around in the wires, I didn't want an unpleasant surprise. Guess what? They were flat. No spark. This is maybe 3 minutes, post-shutoff. I don't see any drain resistors in the circuit (yeah, I know there's something, plus the internal impedance of the caps), so why did these guys drain so fast? Theory: The caps are really bad. The dielectric is breaking down, so they won't filter well, but do leak a lot. Current leakage leads to heat. Heat leads to more leakage, and eventually to a vent - but before that happens, the fuse blows. That make sense? Wish I would have felt the caps...

So here are my questions du jour:
1) What other places should I investigate re the fuses blowing? Is this indicative of a "bad" 6CA7? Or just the cumulative effect of old caps? Or could the V2 12ax7 be doing something bad, or even the V4 12ax7 phase splitter or V3 12ax7 mixer? Eventually, I'll figure it out, but a hint won't hurt me.
2) Is a 6CA7 identical (for my purposes) to an EL34? I.e. can I plug an EL34 into a socket wired for a 6CA7 without modifications, including biasing? This guy has a fixed bias circuit for the moment, and I really plan on leaving that alone, if possible. But EL34's are a lot easier to find than 6CA7's and the selection is better. I haven't done my homework here.
3) This guy is running +440 on the plates of the power tubes. If you were me and trying to rebuild affordably, yet with quality, and tone and reliability overcome cost... what commonly available (GT, Sovtek, EH, JJ, etc.) matched duet would you get for this amp? I've heard good things about the JJ EL34's - and they're fairly cheap. And, I've heard that some brands don't tolerate the high B+ all that well.
4) Anyone know of a good reason not to use a Corcom or equivalent type combo EMI filter / IEC power input as the main power input connector for my rebuild? I've seen several designs do some rudimentary RF filtering on the AC mains side of things, but no one has done this. I happen to have a couple of appropriate size laying around. I figure it'll dampen any RF EMI coming out of this guy (there should be none; that's what the filter was designed to do) OR in (which is what I'm more worried about).

The project proceeds...

Thanks for all your help,
Dennis

[STARDOM]

I bought A TRAYNOR YVM1 for CAN $250 and had $170 of work done on it to anyone considering modifying please do so it will be the best thing you could do for this particular amp. and if possible follow as shown below:
*Add a mid control pot on channel 4 permanently.
*Ground your power cable with 3 prong wire.
*Replace 2 filter caps in the pre-amp/inverter circuitry.
*Rewire pre-amp ground connectors (for the least amount of hum).
*Suplemented power amp filter caps with the replaced pre-amp/inverter filter caps (better low end response and faster response overall).
*Rewire channel 1-2-3 to guitar stock circuitry (instead of mic impedance as it came with for pa system circuitry).
*Added a behringer MICROHD MD400 hum destroyer right after the beginning of each channel input 1-2-3 making it balanced and effectively eliminating hum 100%. (I mean like really dead quiet, turn the volume all the way up and stop playing and it feels like you turn the amp off.
*Designed a new Box similar to a VAN HALEN BOX from birch plywood, imitation leather, new face place with extra wholes punched in with knobs for future mods. new grill, restored Traynor logo.
basically looks brand new.
I encourage everyone to do the same for all vintage Traynor products.
I hope this can be helpfull when you are modifying your head, and all these mods are extremely good for metal players. and other types of music.
Sincerely
Hector
PARTS USED:
TWO 20 PF 500V CAPS.
ONE .1 UF 630 V CAP.
ONE 22/35 V CAPS.
5 1/2W RSISTORS.
ONE .022/400V CAPS.
ONE 6800PF/600V CAP.
ONE 100K POT.

[STARDOM]

Dennis read my mods for the TRAYNOR YVM1 voice master
hector

[gazinkus]

Hector:

Thanks for the response, and yes, I did read it. Much of what you mentioned, I planned to do anyway (except for the part about making it look vintage - I'm building a custom amp here off a vintage base; it's gonna look custom.). I'll have to look at those Behringer hum reducer thingies and see what they do. Mostly, after looking at the way the Traynor is wired, I plan on rewiring/recapping/general cleanup. It's pretty much a mess by modern-day wiring standards (ground loops everywhere, high Z signal leads floating around without shielding, no hard ground anywhere, etc.) I plan on using shielded wiring in at least the first couple stages of each of my preamps, maybe more, possibly DC heaters, definitely star-grounding, definitely 3-wire power input with hard grounding, plus a total cap job on all the electrolytics and some of the rest, etc. I don't like hum at all - I even go to active and other hum-free pickups on my single coil guitars; I don't see the point in having Tone From The Gods if it is masked by hum. And, if I'm gonna pay to rework this amp into my dream amp, it's gonna be done so that it doesn't hum, that's for sure.

Dennis

Originally Posted by STARDOM

Dennis read my mods for the TRAYNOR YVM1 voice master
hector

[Enzo]

About the only thing that would light up your EL34 is a bad EL34. Change those tubes.

[gazinkus]

So you don't think the electrolytics alone could cause such a problem. Makes sense. Just to be safe, I have new electrolytics AND new tubes coming. I found a source (finally) for the 2-in-a-can electrolytics (JJ 50/50, 500v versus the Mallory 40/40 450v), and I won an ebay auction for matched quartet of EL34B's (I figure to have a spare set if they work well...or 4 throwaway "test" tubes if they don't). Between that and my already planned mods to the input power section and a few hum-corrrection changes, I should be ready to fire it up again next weekend or so (I'll be out of town for a few days, plus I have to wait on the stuff to arrive).

I guess I'll also have to dig my old government surplus tube tester out of storage, clean it up (probably replace the caps!), and learn how to use it. And, it probably won't hurt to bring the 'scope into my office from the garage where it has been gathering dust. Damn. I'm really getting into this.

Dennis

Originally Posted by Enzo

About the only thing that would light up your EL34 is a bad EL34. Change those tubes.

[Steve Conner]

In my experience, you can get the hum from a push-pull tube amp down below the noise floor, without resorting to DC heaters and such like. So I hope once you've replaced bad capacitors and cleaned up the mess, there shouldn't be any great hum problems.

Single-ended amps are a different story.

[gazinkus]

Thanks to Steve, Enzo, and Hector for comments and suggestions so far. I plan to keep up a running commentary and Q&A session as I continue rebuilding this amp. Your continued guidance is appreciated.

Status: I returned from out of town (wayyyyy out of town... Maine) early, early this morning and resumed my amp rebuilding later this afternoon. The caps I had ordered had arrived earlier in the week, and the tubes arrived in the afternoon.

First project was removing the old Mallory aluminum-case 2-in-a-can caps. The Traynor folks had soldered all of the ground tabs to the steel chassis as a mounting/grounding mechanism. With four solid mounting points and only a couple small soldering irons (A Weller solder station, and a Weller 140W gun), I didn't have enough heat to melt any one corner long enough to get solder out and wiggle the tabs loose. So, I had to break out the Dremel tool and cut off the tabs. Eventually, I got the cans off, and went back with the gun to remove the tabs from the chassis. I never got all the old solder off, but I made it acceptable-looking. Then I drilled the chassis for the new cap mounts and installed the new caps. Then I wired the new caps in place, replacing the old chassis-grounding with a star-ground connected to the green ground from the input power module. I also replaced the old 2-wire accessory socket with an IEC-type socket, and removed the whole ground-switch circuit, putting the power switch and fuse both in the hot leg headed to the power tranny. I replaced the blown fuse with a 2A slo-blo, figuring that if I had any problems, it would blow pretty quickly.

I put two of the EL34B matched quartet into the power tube slots, and replaced V2, V3, and V4 with 12ax7a's - including 2 of the ones originally in the unit. I left V1 open, as it is not being used in the current preamp configuration. And, holding my breath, I fired that sucker up.

No problems on standby. All tubes lit up. No smoke. And I hit the standby switch. A little hiss... but no appreciable hum. Already an improvement! I worked the preamp and master volumes. Some pot scratchiness... I guess that's to be expected from a 40-year-old amp. They're probably headed for replacement, anyway, but I'll try to clean them later. No crackling or popping, though, as before. So far, so good.

So, I plugged in the Frankenstrat and proceeded to wank away at fairly low volume. Not bad. No appreciable hum, no weird noises - other than those produced by my (lack of) virtuosity. I dimed the preamp volume. Hum grew. OK, I expected that. But not objectionable. I cranked the master volume (bear in mind that my cabinet of the moment is a dual-5.25" 75W monitor cabinet... it can handle the load but is never going to sound good, and I'm not sure I want to really press the issue in re cone excursion). Hum grew with volume. But everything worked. So far, success.

Tone was pretty brittle and without character. But, hey, we're dealing with a PA amp here. And a single-coil guitar. And 5" speakers. And cautious volumes. I suspect things will change dramatically when I start modifying the tone stack and the preamp structures. And break out a real cabinet, and the PRS. And throw caution to the wind. But one must crawl before one walks.

So, I will continue to work on my hum improvement techniques - star grounding, shielding of early-stage signals, replacement of various caps. I have observed that my circuit is NOT the same as either of the published circuits I have found (a '66 and a '69). For instance, there is no choke in the B+ circuitry. So, I'm going to have to map it out and document the differences. And, also in the short-term works will be the renovation of the V2 circuits - currently two PA-type inputs - into a single Plexi channel with a boost stage. And, I need to drill/punch the chassis for two more 12ax7 slots.

But I'm encouraged, so far.

Thanks,
Dennis

[gazinkus]

OK, things have been slow for a couple weeks as I have been tied up with life its ownself.

New developments on the Traynor front:

1) While mapping out the schematic for the YVM-1 that I'm starting with, I have discovered that the output impedance selector switch is wired differently from either of the two schematics I have found (1966 and 1969). Those two have a triple-tapped output transformer and a 16/8/4 ohm selection capability on the 3-position slide switch. Mine appears to use a different (double-tapped) tranny but the same switch, and the switch appears to be wired for 4/4/2 ohm. The back panel is actually marked 4/ /2, but the middle position clearly selects the same thing as the 4 ohm setting. After thinking about it a bit, this actually makes sense. There are 4 output jacks (at the moment), wired in parallel. This is a PA head. Speaker cabinets are typically 8 ohms. You're gonna use at least two. Doesn't make sense to have a 16 ohm setting, and only marginal sense to have an 8 ohm setting. And, they probably had a boatload of those 3-position slide switches sitting around from earlier production runs.

2) Also, to my surprise, the primary output jack is a shorting type. If nothing is connected the output tranny is shorted out. Strange (to me). On top of that, there is a 180 ohm, 15 watt resistor wired across the output jacks. What possible purpose could this have? If the purpose is to have a load on the tranny, the resistor isn't in effect unless there is a speaker cord plugged in, and then there's a speaker out there - unless it opens up. Is there a problem in these designs with running with no load on the tranny? The combination of these two items makes sure there is always a load on the tranny - either a speaker, a short, or a 180 ohm resistor.

3) I obtained a 4x10 cabinet of suitable size and configuration (8 ohms though) to crank up the amp. And did so. Loud as hell, even on about half volume. After a while, it appears that one of my preamp tubes blew. They're old, so that isn't unexpected. But, I went in and checked diodes and all power supply caps (especially the bias caps I haven't yet replaced), and they all seem good. So, I have to replace a 12ax7 before cranking it up again. No biggie.

4) Given my 4x10 cabinet with 8 ohm impedance, I'm presuming that it has 8 ohm speakers. My only other choice here is to wire the speakers in parallel and end up with 2 ohm cabinet. Or wire it as a stereo, 4 ohm cab and maybe only use half of it. Despite the amp's apparent capability to handle this, 2 ohms gives me the heebie jeebies. Anyone have any ideas here? The output tranny is a Hammond A1333, which I haven't found specs for. Does this 2 ohm thing bother you guys as much as it does me?

Thanks,
Dennis
Wimberley, TX

[gazinkus]

Latest issues:

1) I swapped out tubes, one by one, until I found the one that would start misbehaving after about 2 minutes of applied B+ . It was the phase splitter tube. Replaced it, and all is well.

2) I measured the power supply voltages. It appears that my setup is a bit hotter than the 66 and 69 Traynor setups, with +445 max, and +410 for the preamp anode supplies (!). I think I've determined that this will drop to about +367 (I need to up the wattage of a PS resistor...) with my anticipated load. Here's the issue, assuming that I indeed get a +367 out of this rig prior to the anode resistors on my 8 planned preamp stages (1/2 12ax7 each). I believe that the anodes want to see a specified voltage in order to behave per the design(s) I am emulating. Accordingly, if I have 367 volts at the supply side of the anode resistor, and I want to get to, say 200 volts at the anode itself, I need to drop that 167 volts across the resistor. Ohm's law tells me how big the resistor needs to be, if I want the anode current to be the specified value. So, even if the design I am emulating has a 100K anode resistor there, to get the anode voltage and current of the original I might need to put in a 200K resistor there, because I have a hotter supply than the emulated circuit. Right? Right?

Thanks to all you experts...

Dennis
Wimberley, TX

[Steve Conner]

Hey Dennis, glad you're having fun :-)

It's probably not a good idea to change the plate resistors like you said, because it also affects the AC gain of the stage.

If you look at your amp's schematic, you should find that the B+ rails to the various preamp tubes have RC filters on them for decoupling. The "R" is usually 10k or 27k or whatever. To lower the plate voltage of a preamp stage simply make this R bigger.

On a similar note, you ought to check the actual line voltage at your location relative to what the amp's voltage selector is set to. I hear that line voltages have got hotter over the years in the US, and maybe other countries too. (I live in the UK, which is nominally 230V now, but my line voltage actually measures between 240 and 250V, so I crank those voltage selector dials up as far as they'll go.)

steve

[gazinkus]

Thanks, Steve...

Yeah, it's a hoot so far. I'm finding out just how little I know about something I've been around all my life.

It's probably not a good idea to change the plate resistors like you said, because it also affects the AC gain of the stage.

OK. I can buy that. And, it's also why I asked the question. But, it begs another question - what the heck is the difference in changing the resistance between some B+ value and the anode of a tube at one resistor or the other - they're essentially in series. And, by adjusting the 8 individual anode resistors, I can specifically tailor the anode voltages to what the designs seem to expect - at the tube.

If you look at your amp's schematic, you should find that the B+ rails to the various preamp tubes have RC filters on them for decoupling. The "R" is usually 10k or 27k or whatever. To lower the plate voltage of a preamp stage simply make this R bigger.

Yeah, there's a 10K between the phase splitter offset voltage and the feed voltage for the anodes. In the Traynor, this latter voltage goes straight to the 470K anode resistors for all of the parallelled input stages and a 220K for the mixer stage. I'm replacing all of that with a Plexi-type channel and a Dumble-type channel. Best I can tell, the Plexi wants to see something like 191 or 161V at the plates, depending on the stage, and the Dumble wants 206 to 229, depending on the stage. I've calculated the anode current that each stage wants to see per the designs I'm ... uh ... borrowing. It totals about 7.3mA.

My first thought was to just double that 10K to 20K and drop some volts there, and I guess that's what you're suggesting. But given that the Traynor supply seems to be giving me 410 (I measured it with a crappy analog meter and the existing PA circuit with one 12ax7 gone), I'm wondering what it will really give me when I start loading things.

Has anyone just dropped the V's with zeners? Sure would take a lot of guesswork out of things. I could rip out that power resistor, put in a couple suitable zeners and caps, and end up with EXACTLY the voltage I want at the heads of the anode resistors...

On a similar note, you ought to check the actual line voltage at your location relative to what the amp's voltage selector is set to. I hear that line voltages have got hotter over the years in the US, and maybe other countries too. (I live in the UK, which is nominally 230V now, but my line voltage actually measures between 240 and 250V, so I crank those voltage selector dials up as far as they'll go.)

Well, I'm kinda lucky there. This is a single-supply model with no voltage selector. US only. So, whatever I do has to be accomplished in the secondary. Just for grins, I just checked the line voltage AT the power jack of the amp... 122 VAC. The '69 schematic specs 115VAC. Mine isn't the same as the '69, but isn't all that different, either.
Dennis

[Enzo]

Unless this is a finely tuned machine, I might suggest worrying less about the voltage. It is real typical for voltages to be a lot higher in old amps now than they were back when, yet they still sound OK.

But given that you want your target volts:

what the heck is the difference in changing the resistance between some B+ value and the anode of a tube at one resistor or the other - they're essentially in series.

They are in series for DC, but not for AC. That is the purpose of the decoupling cap. As Steve said, changing the plate load resistor will alter performance of the stage. REmember too that for a given B+ node, there might be four triodes - or more - running off it. All their currents add up. So altering the current through one triode will affect the voltage drop through the common resistor. The resistors beyond the decoupling cap are not part of the load.

Has anyone just dropped the V's with zeners? Sure would take a lot of guesswork out of things. I could rip out that power resistor, put in a couple suitable zeners and caps, and end up with EXACTLY the voltage I want at the heads of the anode resistors...

But the zeners need a certain amoint of current flowing to zene properly, won't they? That will likely be a lot more than your 7ma.

There is a relationship between your B+ and the mains voltage too. Roughly, for every volt of change in the mains, there will be about 4 volts change in the B+. A 5v change in the mains can mean a 20v change at the rectifier. And that is one reason there was the notation on the old Fender schematics that said: voltage readings may vary +/-20%. I wouldn't sweat the difference between 191v and 206v.

[gazinkus]

Unless this is a finely tuned machine, I might suggest worrying less about the voltage. It is real typical for voltages to be a lot higher in old amps now than they were back when, yet they still sound OK.

Well, I dunno about how "finely tuned" it is/will be, but it seems that if I'm shooting for a reasonable Plexi / Dumble imitation, it might be appropriate to emulate the anode voltages as well as the component values of the preamps. And the fact is, at the moment, I have too much oomph as compared to those designs. I need to reduce the oomph somehow.


They are in series for DC, but not for AC. That is the purpose of the decoupling cap. As Steve said, changing the plate load resistor will alter performance of the stage. REmember too that for a given B+ node, there might be four triodes - or more - running off it. All their currents add up. So altering the current through one triode will affect the voltage drop through the common resistor. The resistors beyond the decoupling cap are not part of the load.

Doh. It just dinged on me (I'm slow sometimes but eventually I catch on...). OK, so somehow I have to drop the B+ at the anode resistor head to something approximating the original design value, so that I can maintain the anode resistor value, therefore the gain structure of that original design.

But the zeners need a certain amoint of current flowing to zene properly, won't they? That will likely be a lot more than your 7ma.

Worse yet, since I'm likely to need 3 or so different B+ values for the preamps, each zener might only be handling 2mA or thereabouts. I didn't think about Iz.

There is a relationship between your B+ and the mains voltage too. Roughly, for every volt of change in the mains, there will be about 4 volts change in the B+. A 5v change in the mains can mean a 20v change at the rectifier. And that is one reason there was the notation on the old Fender schematics that said: voltage readings may vary +/-20%. I wouldn't sweat the difference between 191v and 206v.

I got the 191V value from a website showing the measured specs from a Marshall JTM45 (191 and 161, actually, at the anode, depending on the stage). I got the 206, 209, 227, and 229V values from the Dumble Clone website. I also have the power supply values for the Dumble Clone (323 and 361), but don't really have them for the Marshall. So, for the Dumble, anyway, it's easy to calculate the anode currents.

It's beginning to look like I need to redesign the anode power supply. Right now, I have a 445V B+ (for the audio tranny center tap) feeding a 4.7K and cap that supplies the phase splitter offset voltage. Then that feeds a 10K and cap for the anode supply. I measured 445, 440, and 410 at the three points with the old loading. So, I guess I need to figure out how much current is going out to the phase splitter. If I'm losing 5V across a 4.7K, that sounds like about 1mA to me. And that means I'm starting with 440V that I need to drop and filter for each of my anode supplies. I know I need 7.3mA total for the 8 preamp stages. So, to get to my first value of 361V (or thereabouts) that means... about 11K for a filter resistor. Right? Hmmm, that ain't all that different from what I have now (10K). Maybe I just leave that alone and branch off of it for the lower anode voltages I need. Comments?


By the way, thanks for the help.

Dennis

[Enzo]

Well, if those 8 stages are not powered off the same B+ node, then consider too that your 7ma is only at the power supply end, as each stage taps off the B+, then the current through the later node resistors is reduced by that much.

[gazinkus]

Originally Posted by Enzo

Well, if those 8 stages are not powered off the same B+ node, then consider too that your 7ma is only at the power supply end, as each stage taps off the B+, then the current through the later node resistors is reduced by that much.

Well, after considering your advice and contemplating upon it... and, ummm... borrowing... an idea from the Hoffman Amps version of a Plexi circuit with boost, I note that the Hoffman B+ supply has a 10K feeding a cap and the higher-voltage anodes, followed by another 10K and cap stage feeding the lower voltage anodes. If my calculations are correct, the first 10K - at the load I expect (and evidently the load that Hoffman expects) will draw the B+ down to around 365 (close enough to 361 that the 3rd and 4th stages of the Dumble Clone design wants, and close enough to what it looks like the 3rd and 4th stages of the Hoffman design wants), and another 10K will drop another 40 volts and get me down in the 325 range for the lower voltage first two stages of both preamps. So, essentially I end up duplicating what Hoffman did for the last two stages of their power supply and look to make both preamps have near the expected voltages.

Right now, I only have the first 10K and cap - Evidently Traynor liked to run all of their preamps hot. But given that the YVM-1 only had 2 stages - 4 paralleled 1st stages mixed by a 2nd stage, I guess it makes sense. If you want only have two stages, you might want all the oomph you can get.

Thanks again, Enzo...

Dennis

[Pentamonium]

[QUOTE=gazinkus;28407]
So you don't think the electrolytics alone could cause such a problem. Makes sense. Just to be safe, I have new electrolytics AND new tubes coming. I found a source (finally) for the 2-in-a-can electrolytics (JJ 50/50, 500v versus the Mallory 40/40 450v), ...

Where did you find those capacitors ?

Thanks
Stefan

[gazinkus]

Stefan:

I used Antique Electronic Supply, http://www.tubesandmore.com/ . Go to Capacitors, Multisection, JJ Electronic LCR. I think they're the 3rd item down. $8.75 ea (US) without the mounting rings.

I had to work pretty hard to get the old Mallory's removed from the Chassis, then drill holes for the mounting rings. Now, at a cursory glance anyway, it looks like the JJ's have been there all along.

Good luck,
Dennis

P.S. Yeah, it ended up being more than just the electrolytics. Turns out that several tubes were bad. The phase splitter tube would wait a couple minutes to warm up before it would start to flake out, and one of the EL34's was bad, as well. But everything else behaved once I changed all that out.

[Enzo]

If I want to remove a bad multican cap, and I don't want it to be saved for some odd reason, I often just cut the twist tabs off at the chassis with flush cutters, rather than trying to straighten them and melting solder so I can pull them out through the mounting slots. You can also just grab the twisters with pliers and work them back and forth until they break off.

[gazinkus]

Originally Posted by Enzo

If I want to remove a bad multican cap, and I don't want it to be saved for some odd reason, I often just cut the twist tabs off at the chassis with flush cutters, rather than trying to straighten them and melting solder so I can pull them out through the mounting slots. You can also just grab the twisters with pliers and work them back and forth until they break off.

In this particular case, the tabs were soldered to the inside of the chassis, pinning the can against the top side. I had a heck of a time getting enough heat applied to get the solder loose. I ended up cutting the tabs with a Dremel tool, freeing the can, then individually attacking the solder points with the biggest gun I had (350W). I didn't get things as clean as I would ordinarily like - but I got the tabs off and most of the surplus solder. Traynor must have used the Soldering Iron From Hell to put them on in the first place.

Dennis