6V6 Iron With EL84 Tubes?

[neddyboy]

This is a follow-on from my 2x12AX7 preamp build. I'd like to build a power amp to go with it using a variation of an old Dynaco power amp schematic (attached). I'm going to jam it all into the same size Bud CU-347 project box that I used for the preamp. I know it's ridiculous, but it sounds like fun.

I have a spare set of Grommes LJ-2 transformers sitting around, intended for use in a PP 6V6 (specs also attached). Is it possible to use these transformers in an EL84 circuit? The Grommes PT puts out 340V B+, but I'd need to drop it to 300V for the EL84s, right? FYI I'd like to use an EZ81 (or similar) rectifier rather than an octal for size reasons. I assume I'd have to drop the voltage before the rectifier tube with a big ol' resistor, but is 40 volts too much to dissipate? And I'm totally out of my depth with the OT -- will it work with EL84s? Any other issues I'm not seeing?

Thank you in advance!
Neddy

[Diverted]

You could use a string of 10V 5W zener diodes from the HV center tap to ground, to knock down the voltage. They’re easy to use, stick them on a terminal strip. Keep them off to the side of the chassis as they throw off some heat. Try 3-4 of them in series; that’ll get you down around 300. Striped end points to ground.

[Diverted]

You can use a 6v6 output transformer for EL84s and vice versa. The only difference will be a bit of a difference in biasing/watts as the primary impedance specs for 6V6 vs EL84 are slightly different (8K vs 6.6K I believe).

[66Strat]

Not enough current capacity in the 6.3 volt winding. Each EL84 wants 760 milliamps, the 12AX7 wants 300 milliamps, and the EZ81 wants 1 amp. Add it all up, and you have 6.3 volt heater current demand of 2.82 amps. The 6.3 volt current supply is limited to 1.4 amps. I wouldn't worry too much about plate voltage. Keep plate dissipation less than 12 watts, and you should be good to go. You can play with the circled dropping resistor value to get your target B+.  The output transformer is fine.

[dude]

340 volts is no problem for EL84s, l’d skip the zener or resistor to lower voltage. 18watt amps run all day with OT 8k impedance, 360v B+ and EL84’s. As far as the EZ81, why go through the trouble to reduce the socket, maybe use 5Y3.


If you’re PT is an issue, use diodes rather than a tube, voltage will be higher but even 360 is fine for EL84s. If you want lower voltage for tone, use the power resistor or zenner’s off CT like diverted suggested

[neddyboy]

Hmm, OK! Dude, I was originally planning on a diode rectifier and have a bridge ready to go. I only thought about going back to a tube to drop some voltage, so SS rectified it is. Amazing that the EZ81 draws that much! One question tho -- this is looking a lot like the power amp section of an 18 watt Marshall, and IIRC the 18 watt circuit doesn't play well with SS rectifiers. Have you heard that?

[66Strat]

Hmm, OK! Dude, I was originally planning on a diode rectifier and have a bridge ready to go. I only thought about going back to a tube to drop some voltage, so SS rectified it is. Amazing that the EZ81 draws that much! One question tho -- this is looking a lot like the power amp section of an 18 watt Marshall, and IIRC the 18 watt circuit doesn't play well with SS rectifiers. Have you heard that?

You're still underwater current wise. Two el84s draw 1.52 amps add in the AX7 and you're at 1.82 amps. The PT 6.3 volt winding is limited to 1.4 amps. Why not build the amp as designed?

[neddyboy]

Dang, yeah, I misread your post earlier. I guess it's a no go. Why not build it as shown? Bad reasons! I'm unemployed and trying to build this with parts on hand, and the only chassis I have is a 8"L x 5"W x 3"H project box. I have to admit too that I have an unhealthy fascination with miniaturizing. Anyway, once I get the PT and OT in there, the octal sockets and tube diameter make it very challenging. The 9-pin tubes allow a lot more clearance. I even have a 6.3 volt transformer, but there's be nowhere to put it on this tiny box! I have one more Bud box the size of a large guitar pedal, so I could build a separate power supply box with the PT and the filament transformer, but that seems silly for a guitar amp.

[shooter]

 
be creative;
http://sluckeyamps.com/smoky/smoky.htm

3lb coffee can and your box 
I built a wooden box under a metal box, worked well

[66Strat]

Dang, yeah, I misread your post earlier. I guess it's a no go. Why not build it as shown? Bad reasons! I'm unemployed and trying to build this with parts on hand, and the only chassis I have is a 8"L x 5"W x 3"H project box. I have to admit too that I have an unhealthy fascination with miniaturizing. Anyway, once I get the PT and OT in there, the octal sockets and tube diameter make it very challenging. The 9-pin tubes allow a lot more clearance. I even have a 6.3 volt transformer, but there's be nowhere to put it on this tiny box! I have one more Bud box the size of a large guitar pedal, so I could build a separate power supply box with the PT and the filament transformer, but that seems silly for a guitar amp.

Been there done that on more than one occasion. I was a victim of corporate "right sizing" on 4 occasions. I hope that you have better fortunes soon. With the box that you have, I believe that you may have enough room to build a "tweed" style box. Input and controls mounted outside on one 3 inch side with the tubes mounted to the opposite 3 inch side. Transformers mounted outside on the 3x5 with the circuit card mounted inside. Mount the completed chassis in a mini head box. It would be a tight squeeze. But, I bet you can do it. It's worth scribbling out some layout diagrams to try for size.

[shooter]

 

"right sizing"

I got sold once and down right    the last time
Tubes are good therapy 

[neddyboy]

You guys are inspiring me to heights of "creativity" aka silliness. FYI 66strat, this build is a discrete power amp with (probably) no controls, just a power switch, a fuse, and a pilot light. It's meant to pair with the preamp I built recently. I'm going for a component-style setup with 2 separate chassis for no good reason other than having the parts already and being bored. Maybe I will build a separate power supply on that last project box!

That raises a new question: I'd likely make my own umbilical out of transformer-lead type wire. Do any of you guys know of any issues with running HV and heater parallel in an umbilical? Ah jeez, would I have to twist everything for the length of the umbilical?

[neddyboy]

Shooter, I built a spring reverb unit using a Trader Joe's peppermint bark tin as a chassis. I'd show you a picture, but it never worked well and became the donor for the preamp project. I love that purple guy too!

[66Strat]

You know, there's nothing etched in stone that says that you must build using a metal chassis.

[kagliostro]

About heater consumption it was told that EL84 requires more current than 6V6

I'm wondering which will be your B+ with a 680V AC CT transformer



340V AC x 1.4 (Solid State rectifier) will result in 476V DC B+

assuming you solve the heater consumption problem (an additional heater transformer) you can drop B+ a bit using a Vacuum Tube Rectifier, but I think you'll obtain around 440V ...... it will be better you use 6P14P-EV tubes (OK till 500V) instead of EL84 tubes (OK till 300V)

K

[neddyboy]

OK, I decided to bail on the EL84 substitution. Attached is my attempt to arrange the components for a full 6SL7/6V6/5Y3 circuit on my chassis without using a separate power supply chassis. Pretty tight, but it seems possible. Clockwise from top right, 5Y3, 6V6, 6V6, 6SL7. I chose to put the OT closer to the PT so the tubes could be farther away. I may have the option of putting the OT inside the box, like it was on the Grommes. I plan on installing a steel bulkhead inside the box to isolate the power section. Given that, would the OT be better shielded inside the box or outside? Any other comments/suggestions on placement would be most appreciated!

[shooter]

maybe move the OT to the back of the chassis

[neddyboy]

Thanks Shooter! You think mounting the OT outside the box on the back would be quieter than inside the box with a bulkhead? I'm resistant to the idea, as the OT & PT at the back might affect the balance the box and make it likely to tip over backwards. These things will sit on top of a speaker cabinet, and I'd like to keep them there! Do you think the benefit would outweigh the potential problem?

[shooter]

If the amp tips over you're playing to loud 

the OT is noisy, you want it away from the preamp AND "decoupled" from the PT
putting it inside changes planes with PT but puts it even closer to the preamp tubes.

If you're like most, down the road it'll become something else, so where you mount it now will gain you knowledge for the future 

[HotBluePlates]

...I'm wondering which will be your B+ with a 680V AC CT transformer



340V AC x 1.4 (Solid State rectifier) will result in 476V DC B+ ...

476vdc with no load, which is important for filter cap ratings.

But the PT also has winding resistance, and the rectifier has voltage drop.  The PT in my Vibro Champ has a loaded winding voltage of 340-0-340vac, but after the 5Y3 the voltage at the 1st filter cap is 388vdc.  In this case, the B+ winding resistance is 378Ω, which lowers the B+ voltage right along with rectifier internal resistance.

If it weren't for other constraints, the OP could add a resistor between the B+ winding & each rectifier plate to help knock down charging current & B+ voltage.  It would probably be wise to add a solid-state diode between the resistor & the actual tube rectifier plate, to guard against future rectifier failures.

A lot of modern replacement transformers seem to have much lower winding resistance than the actual vintage parts, which plays a factor in the higher B+ output we keep getting.

[neddyboy]

HotBluePlates, I've decided to be conservative and build the 6SL7/6V6 circuit. I'd have to buy almost nothing, and if I use a diode rectifier I can fit it all onto the chassis. Would I see higher voltages with diodes than with a tube? I don't know anything about diode vs tube past the basics. My idea was to use a diode bridge and a sag resistor (rather than a tube) as they'd be easier to squeeze into the chassis next to the PT, but I'm not sure if this is feasible. Thoughts?

[shooter]

here's a good read on reciters, he also covers the rest of the PS

http://www.valvewizard.co.uk/bridge.html

[neddyboy]

I'm still exploring some wackier options on this, and I have a couple more arcane (for me) questions. Rest assured, I do try to track this stuff down before I ask you guys.

First, if I were to build a separate power supply box including the caps, rectifier bridge, and a choke, I could then send B+ voltage through a hard-wired umbilical. IIRC it's not good to run AC filament voltage alongside the DC. Do you guys know anything about this stuff? I could A. install a 6.3v filament transformer in the power amp box, or B. rectify the 6.3v AC from the PT before sending it down the umbilical. My vast parts bin includes a Hammond filament transformer, so no added expense. Any thoughts or OMG-don't-do-thats?

Second, and related, could I place both the filament transformer AND the OT in/on the box shown above without adding a lot of noise? I've done some reading and I think I get the basics of transformer alignment, but my situation doesn't come up in anything I've seen! My preference, mostly for aesthetic reasons, would be to mount one transformer inside the box like the layout attached. I've shown the filament transformer mounted inside the box aligned vertically on the the side wall, and the OT on top. Advisable?

[sluckey]

if I were to build a separate power supply box including the caps, rectifier bridge, and a choke

My Magnatone M10-A amp does just that. There is no problem having B+ and filaments in the same cable.

     http://sluckeyamps.com/magnatone/magnatone.htm

Here's another similar project. PS and PA on one chassis, preamp on another chassis with connecting cable that carries B+, filaments, and signal. No problems.

     http://sluckeyamps.com/vox/vox.htm

[neddyboy]

Thanks Sluckey!

[neddyboy]

OK, checking back in! I decided to build a separate PS box using a Classic Tone 5E3 PT I had. It's about the same as the Grommes PT, but with slightly higher voltages. I'm planning on 6 conductors in the umbilical: 3 @ high voltage, 2 @ 6.3v, and ground. The 5E3 circuit runs high voltage to the OT direct from the rectifier, which would require another conductor in the umbilical -- why couldn't I just tap into the high voltage from the other end, within the power amp chassis?

[sluckey]

You only have 3 filter caps so you only need 3 high voltage wires. Think about it.   

[neddyboy]

Oh, I do. Sometimes I don't trust my thinks though! Follow-on question: as you know, I used the Tube Pedal schematic to build my preamp. I'm planning on using this with my next project, a 5E3 power amp in a tiny box. Can I assume that the only tube I'll need in the chassis with the output tubes would be a 12AX7 to act as a phase inverter?

[shooter]

yup, unless you use a transformer PI, but then it'd be a Gibson 

[neddyboy]

Thanks shooter! I've built two tweed Deluxes in normal chassis, but this project is challenging my weak understanding of theory, AKA actually knowing how stuff works. Watch this space for more questions...

[neddyboy]

Next step, the power amp! I'm putting it into another 5" x 8" x 2" chassis. I have 8" of space to use for the OT, the 2 power tubes, and the driver/PI, and I'm unsure about placement. The OT will be on top at one end, oriented across the 5" width. I'll put a bulkhead inside, between the bottom of the OT and the tubes. Here's the question: should all the tubes be as far as possible from the OT, even if it means the driver/PI is close to the the power tubes?

[shooter]

Next step

the PS tests out good without a load 

I've put power tubes < a tubes width from the OT no problem.  the Pi isn't the sensitive type like V1, but should keep it's distance from the OT

[neddyboy]

I'll be testing the power supply tomorrow, fingers crossed. I had to wait for a fuse holder to arrive. I will keep you posted!

[neddyboy]

I brought up the power supply on a variac, and the unmarked probably-a-5Y3 rectifier was indeed a 5Y3. The voltages I got are A: 482, B: 486, and C: 450. Please bear in mind that I replaced the 5k dropping resistor with a Fender-correct Hammond choke. The voltages seem too high even for unloaded. Opinions? How would I address this if it is an issue? An extra resistor before the choke?

[sluckey]

Don't do anything until you have the proper load connected.

[neddyboy]

Will do!

[neddyboy]

Hello all! Since I posted last, I finally got the 500v aircraft connector for the umbilical, got the umbilical built and tested. I'm bummed that I ran out of room for a pilot light on the PS chassis. I even tried an LED, and there's just no room. I'll rely on the pilot light on the power amp chassis, and they'll never be disconnected. I finished the power amp section today and got the voltages measured. They're weird: A: 20v B: 335 C: 350. That 20v on A was up in the range of the other two before I connected it to the OT. Why would it drop like that? It's a 60-year-old transformer, but it worked when I stuck it on a shelf!

[shooter]

That 20v on A was up in the range of the other two before I connected it to the OT.

you mean hooking it to the PA - with tubes?  If so you might have a BIG problem with the PA sucking all the current

[sluckey]

They're weird: A: 20v B: 335 C: 350. That 20v on A was up in the range of the other two before I connected it to the OT. Why would it drop like that? It's a 60-year-old transformer, but it worked when I stuck it on a shelf!

If your node A is connected to the OT I would expect the other nodes to be reading low also. You must not have your nodes labeled like we normally see them labeled.

[pdf64]

Does your parts bin include a fuse for the HT?
Either there’s a bad connection or, as noted, a lot of current being drawn.

[neddyboy]

Yes, there's a fuse in the PS. My skill level demands fuses!

I tested the PS unloaded and all was well, so the error must be in the PA. I also suspect a bad connection, or should I say, a swapped connection between the 3 nodes. I'll be re-re-checking them today. Sluckey, I'm referring to them as A B and C based on Weber's 5E3 schematic. I didn't even know they were called nodes until now! How should I be labelling them?

Shooter, what would be "sucking all the current"? This is completely new to me. IIRC the voltage on A without tubes installed was correct. When I brought it up on the variac with tubes and speaker, nothing catastrophic happened -- no smoke, no blown fuses, no hot transformers -- but when it got up to 115v I did see 2 pops inside one of the 6V6s before I shut it down. Shorts maybe? Last, although I did have the preamp connected and a signal going in, I got no audio at all through the speaker.

[shooter]

Shorts maybe?

that's what sucks current, Ohm says so 

[sluckey]

Sluckey, I'm referring to them as A B and C based on Weber's 5E3 schematic.

That's the commonly accepted way to label the nodes. However, it is impossible to only have 20v at node A and still have 300V at the other nodes. I suspect the OT is not connected per your schematic. You most likely have the OT connected to Node C in order for the other two nodes to still have about 300V present. It would be very easy to miss-wire since you are using an umbilical cord. Should also be an easy fix.

[neddyboy]

Sluckey, you are the Dr. House of amplifiers. I'm very embarrassed to say that I did swap the A and C high voltage wires. I screwed myself by trying to be too organized. Trying to deal with A-B-C-Orange-Yellow-White-Pin-1-Pin-2-Pin-3 broke my brain!

I definitely don't understand why swapping the two HV leads would cause such a power imbalance tho. The 3 nodes aren't that different in voltages. Anyway, that's corrected. With tubes installed, I got A=386, B=386, and C=328, and no shorts. Still high, but at least less weird.

FYI the 22k dropping resistor got so hot it changed its value! Not really, but it did change red-red-yellow to black-black-grey. It still measures correct but I'll replace it. Any thoughts on the voltage measurements? No voltage drop across the choke?

[sluckey]

Sluckey, you are the Dr. House of amplifiers.

No, I'm the Ice House man! 

FYI the 22k dropping resistor got so hot it changed its value! Not really, but it did change red-red-yellow to black-black-grey. It still measures correct but I'll replace it. Any thoughts on the voltage measurements? No voltage drop across the choke?

That's what happens when you try to shove 100mA through a resistor that's only expecting 5mA.

[shooter]

I definitely don't understand why swapping the two HV leads would cause such a power imbalance tho.

surf up some "understanding series circuits"  a good read on ohm's law might also help

[neddyboy]

I am suitably chastened! I love doing this stuff, but every time I try to learn the fundamentals of electronics my mind turns to mush. I have a mental blind spot with electronics theory, and it doesn't help that it's so boring. But I will try again.

I have to say though, this project (and you guys) have taught me a huge amount. Breaking down an amp into its component parts forced me to think more about structure rather than a simple execution to a preset plan. As I mentioned before, this is my first P2P project, and the first for me without using a layout.

What damage may I have caused by swapping A & C? I'll test the tubes, and will replace the 22k resistor. How can I address the voltage measurements?

[shooter]

I have a mental blind spot with electronics theory, and it doesn't help that it's so boring

I'm the opposite, love electronics because it's mostly logical
humans on the other hand are mostly boring 

[neddyboy]

Yup! I suppose I'm relatively technical, considering all the software, code, etc. I used to deal with, but I always had to force myself to pay attention to the technical details. I was definitely an art major, not an engineering major.

[sluckey]

What damage may I have caused by swapping A & C? I'll test the tubes, and will replace the 22k resistor. How can I address the voltage measurements?

Barbecued resistor is probably the only casualty.

Measure voltage at all B+ nodes. Measure voltage on every tube pin.