DuMar-V6 with AO-29 PT

[CrocMule]

I recently built a Sluckey Dual Marshall 6V6 (SluDuMar-V6?) using iron from a Hammond AO-29 amp.

Before getting into the build, I want to talk about the PT.  It's the one made for a field coil speaker (mine is stamped AO-22111-1), so it has an elevated HV.  I had a lot of confusion about whether it would work out.  A couple people said in old threads just not to use that one.  But I had it, had the time and minimal cash flow for new, so I decided to try it.

I read through what I could find here and on other boards, and I knew people had used that PT successfully by adding a 30W 700R load to the HV CT to replace the field coil load.  But I didn't find any info that I could grasp before the build about where the B+ voltage would end up.  So I'll post what I got.

I tried 3 different rectifier tubes that I had on hand: newer JJ 5Y3S, old Jan Phillips 5U4GB, newer JJ GZ34S.  I didn't try it with SS.

All readings taken with 120 +/- 1 VAC mains. 

PT w/o tubes, no CT load:  HV = 379-0-379VAC, CT to ground = 150VAC.

The following readings taken with a 50W 700R from HV CT to ground.  (The 50W resistor was cheaper than a 30W).
      
5Y3S, 2x 6V6S (biased at 9.8W w/ 120VAC mains), 4x 12AX7:  B+A = 331VDC

5U4GB, all else the same (6V6s biased at 9.8W):  B+A = 367VDC

GZ34S, all else the same (6V6s biased at 9.7W):  B+A = 387VDC

YMMV, but I hope this helps other novice builders who are eyeing AO-29 iron.  Despite the high open circuit voltage, with a CT load it's a usable PT.  The biggest problem I have with it is the 7 lb. weight! 

The OT (AO-20936-3) needs a lower-power amp.  In the DuMar, it gets mushy at high volume.  My original plan was to do this on the cheapcheap, but now that it has roared awake I'm looking at a Hammond 1760H replacement.  Hoping for a big improvement.

I just saw an AO-29 a few days ago for $40 w/o tubes on Denver CL. 

Best,
Russ

[CrocMule]

You gotta admit "DuMar-V6" has a great ring to it!  Reminiscent of an early '70s AMC sedan.  The name's frumpiness belies the amp's true nature (fanged hell-stallion).  It's my third build. 

The empty pine cab is from a Fender Stage Lead.  I had to fabricate a chassis, which was an adventure in itself.  Weight got out of control fast, starting with the chassis -- 5 lbs. of steel with all the holes.  Then the gigantic PT, and the choke that's twice the size it needs to be (had it on hand).  With a Greenback, the finished amp weighs 46 lbs.  Seems like it's heavier each time I pick it up.

I wasn't even sure this amp would work out because of the PT -- but I knew I would learn a lot, so I plowed ahead, not worrying about silly things like weight.  I've started a list of things never to do again in a build.

Sorry for the mess on the schematics -- I'm a pen and paper guy, with hand tremors.  If anybody spots problems in the circuit as shown, please let me know.

I live off-grid with an inverter that doesn't hold a steady voltage.  With light loads, it outputs ~115VAC.  But when the well pump or freezer kicks on, it raises the output to ~120V.  It's never steady either -- always wandering.  So when testing voltages, I have to run the amp through a variac and continually adjust it to keep the output near the target voltage.  Total pain.  The solution is an Exeltech inverter with a steady 117.0 VAC pure sine wave, but the money for that has yet to find its way to my wallet.  So the voltages on the schematics will be a little off from each other.

After the first set of tests with the rectifiers, I figured it would be best to bias the 6V6s to 9.8W at 125VAC (9.0W @ 120VAC), so they won't run too hot when I take the amp out someplace with higher voltage.  That's why there's a discrepancy between the rectifier voltages above and the schematic voltages -- the 6V6s were biased a little cooler when I took the 2nd set of voltages.

[CrocMule]

I don't know why there's -mVDC on the tone stack driver input (V3 p2).  There's -16mA on the Plexi side and -29mA on the JCM side.  Does that indicate a problem?  My limited knowledge of tube amp design is dwarfed by what I don't know.

I was wondering about lowering the value of the power resistor between the B and C nodes to raise the voltages on the 12AX7s.  With the 22kR (and the 5U4GB), they're running ~30V lower than the voltages listed on the Huss/Sluckey Plexi 6V6 schematic.  My (possibly incorrect) calcs show I'd need ~12.5k there to bring C up to around 270V.  Would it be ill advised to try a 15k?  Not sure what that would do to the tone.

No matter which 12AX7 I put in V2 (JCM pre) I get microphonics.  Just the nature of the beast I guess.  I tried silicone rings in the past, and they didn't make a difference, so I didn't bother this time.  I'd rather keep the tube shield on.

I'm wondering if I need to make the DuMar into a head.  That would suck because I've been really wanting a 112 combo.  I had to chop the previous combo down to a head too (the MarVo -- British Dual Lite -- the EF86s I bought were all microphonic, and from what I read it didn't sound like silent EF86s are a thing).

Do dampened noval sockets work vs microphonics?  Little silicone shock mount washers?  They seem like a better idea than rings.  I haven't noticed them for sale anywhere.

The DuMar sounds pretty good to my ears, but I've only played through high gain amps a few times, and only briefly.  This is the first time I've owned one.  I usually play edge of breakup, so this is a new experience.  I feel like a kid -- it's so much fun!

Big thanks to everybody here sharing their knowledge.  And BIG THANKS TO SLUCKEY!  This is my 2nd build of one of your designs.  Your schematics and layouts are gold!

Best,
Russ

[pullshocks]

I had one of those PT's and had similar concerns about the voltage being too high.  After doing all the research it turned out that it was fried....


Hope yours works out

[tdvt]

Thanks for posting all of that. Those voltages look pretty usable, great to know!

I have one of those field-coil AO-29 PTs on the shelf myself. I started down that same path, but never got around to using it.

I thought it might be good for a 5E3-ish project, but that is not happening anytime soon.

They are indeed pretty beastly things. My OT was already open-circuit, so that is not in the picture. Will be interested to know how the Hammond works out.

[AlNewman]

I'd be tempted to remove or reduce the 700 ohm resistor at the center tap and see what kind of voltages you get.  Fender runs up to 450V on 6V6 plates.
You could maybe fit in 6L6 with the same PT, which is more better.
The PI and preamp could definitely sit at higher voltages, so cutting the resistor between the screens and the PI plates in half would make a big difference.

[CrocMule]

I thought it might be good for a 5E3-ish project...
They are indeed pretty beastly things.

I'm not sure I'd try another combo with one of these PTs.  It seems like it's twice the weight it needs to be.  It was an experiment -- I guess I know more now about what my ideal 112 combo will be like.  It won't weigh 46 lbs.

I think it needs to become a small head.  I'll keep the Stage Lead cab for the perfect 112 combo and make a new box for this chassis.   

With the OT from the AO-29, I couldn't really tell much difference between the 3 rectifier tubes in tone or volume.  It gets pretty swampy at high volume.  The 5Y3 might have sounded weaker and the GZ34 more strident, but those could easily be preconceptions as well.  It was strange -- I expected an obvious difference.  I did zero all pots and rebias each time though, so not a very quick or accurate A/B/C comparison. 

I'll keep you posted about the new OT.

I'd be tempted to remove or reduce the 700 ohm resistor at the center tap and see what kind of voltages you get.  Fender runs up to 450V on 6V6 plates.
You could maybe fit in 6L6 with the same PT, which is more better.
The PI and preamp could definitely sit at higher voltages, so cutting the resistor between the screens and the PI plates in half would make a big difference.

I was wondering about changing the value of the CT resistor too.  Thanks for the advice.  They're not the cheapest things, but I'll get a couple likely candidates and try them out.

I have a few PP 6L6 heads already.  I love them, but they're hard on the ears.  I'm looking for something that will deafen me more slowly, or perhaps less quickly.  I'm not a gigging musician, so more volume can actually be more of a curse.  I also have a few old PP 6L6 amps waiting in the wings for restoration.  At this point I'm looking for a 6V6 power section. 

I'll pick up 12k and 15k power resistors too.  I may have a 15k on hand.

Thanks!

[tdvt]

You could maybe fit in 6L6 with the same PT, which is more better.

I think that PT is listed at 140mA somewhere on the drawing.

Not sure if that is enough for 6L6s or not, probably borderline.

[PRR]

something that will deafen me more slowly

Then 3dB (half power) is not even a baby-step. You want more like tenth-power, 3 Watts. Pair of 6V6 at 140V, or pair 6AU6 at 200V with >15Kpp load.

[AlNewman]

What happens if you jump the 700 ohm resistor, and turn your bias as high negative as it will go?  Where are your voltages?

[CrocMule]

What happens if you jump the 700 ohm resistor, and turn your bias as high negative as it will go?  Where are your voltages?

Hey Al, thanks for the suggestion, but couldn't that send some pretty high B+ voltages down the line?  I'm curious, but I'm also pretty cautious.  I have a couple lower value 50W resistors in my cart (500 and 300).  I'll order them Monday.  Should be here mid next week.

PRR also has a good point.  I'm curious about the difference in tone, but I'm not really looking for more volume, so higher voltages might not be what I want in the long run.  What kind of change in tone might I expect with higher voltages?  I'm going to try it because, science, but I may be headed in the wrong direction (for me).

Best,
Russ

[sluckey]

Measure the voltage across that 700Ω. That's how much you can expect the B+ to increase if you remove it. And adjusting the bias voltage to max negative will cause the B+ to rise even more.

[CrocMule]

I measured 43VAC across the 700R CT load (-37VDC).

So I figured it was probably safe to go ahead and follow AlNewman's suggestion and jumper across the 700R (after turning the bias all the way down).

This is what I got with no CT load:

B+A 460VDC
B 457
C 297
D 248

Sluckey, it turns out your original assessment (from a past thread) that the B+ will be ~100V higher without the 700R CT load was correct.  Mine went from 374 to 460, and the bias was still too high (~12W), so B+ still needed to come up some. 

To make calcs easier, I'm going to assume with the bias set correctly the difference between a 700R and no CT load would be 100V.  So to raise B+A from 374 to 400V, I'd need to drop the value of the CT load to ~520R.  I'm just treating it as a simple ratio.  To raise the voltage 26V(26% of 100V), I'd need to drop the R by 26%.  I'm not sure if that's how it actually works, but I'll order a 500R and try that to see if I can get the B+A up to ~400. 

I have a fixation on trying to get this thing up to the voltages on the Plexi 6V6 schematic.  Like I said, it may not be what I ultimately want, but I want to hear it nonetheless.  I'm a noob, so seeing what "this" does to "that" is fun.

Thanks for the help!
Russ

[AlNewman]

Where are you measuring the 12 mA?  Across your 1 ohm resistor?  If so, your bias needs to come up, likely to at least 20mA/tube, which will reduce your B+. 
30 watt resistor seems like an overkill in this case?  You may be able to get away with 5 watts, but 10W should be more than enough.  They should be considerably less expensive.  This way you can order a few different values if needed.  Also, usually common values are less expensive than oddballs. 

[sluckey]

Whoa! Slow down. 460V is already too high for most 6V6s. And your 12mA bias is way too cold. I'm assuming that is 12mA cathode current per tube. So, turn that bias current up to 24mA. This will wake up that amp and also cause the B+ to drop. Do this simple experiment and measure your plate voltage while the tubes are biased at 24mA. Then report back.

Just for reference... My amp had 394V on the plate when the tube was biased at 24mA. This is 9.45 watts idle dissipation. My 6V6 is a 14 watt tube. So, my amp is biased at 67.5% of max dissipation. That's a good place to be IMO.

[CrocMule]

WOOPS!  I meant 12 watts, not 12mA.  I apologize!  The bias was too high -- up around 28mA at the coldest setting on the bias pot.

I also realized that one of my DMMs is failing, even with a new battery.  That's the one I was probably using when I measured the voltage drop on that 700R CT load.  I'll go remeasure.

Thanks!

[sluckey]

WOOPS!  I meant 12 watts, not 12mA.  I apologize!  The bias was too high -- up around 28mA at the coldest setting on the bias pot.

OK. Drop the bias current down to 24mA. Recheck the plate voltage and report.

And tell us if these numbers are with or without the 700Ω.

[CrocMule]

I must have been using the good meter, because I got 43VAC / -37VDC across the 700R CT load again.

OK. Drop the bias current down to 24mA. Recheck the plate voltage and report.

And tell us if these numbers are with or without the 700Ω.

The 460V B+A was without the 700R CT load.  With the 700R CT load, B+A was 374.  I only bypassed the 700R briefly to get the voltages.

The bias pot was already all the way cold, so I couldn't drop the current any further than 28mA.  At ~460VDC, 28mA was as low as it would go.  I'd have to change the bias range resistor to dial the bias down lower.  I'm not going to do that though, since testing voltages without any CT load was academic. 

I just remembered I have a power resistor decade box that I bought a long time ago and never used.  I'll try to to dial in the B+A to 400V with that later -- subbing lower values in for the 700R.  I'll report back.

Stupid question:  What's the difference between plate current and cathode current?  I thought current flows from plate to cathode, so aren't they the same measurement?

Thanks!
Russ

[mresistor]

No cathode current is the sum of plate and screen currents.

[sluckey]

I just remembered I have a power resistor decade box that I bought a long time ago and never used.  I'll try to to dial in the B+A to 400V with that later -- subbing lower values in for the 700R.  I'll report back.

Dial it to 420V also. See if you can get 24mA at either of those voltages.

[CrocMule]

~120VAC mains

CT load = 500Ω          400Ω          300Ω
PV =        393V           407            423
KI =         22.8mA       22.2           21.5
PD =        8.96W         9.04           9.09
B+A =      399V           412            427
B =          396             409            422
C =          255             263            272
D =          212             219            227
E =          197             204            211
F =          203             210            217

I biased to ~9.0W at 120VAC, so it would be slightly less than 70% at 125VAC.

What's the abbreviation for cathode current?  Pretty sure it's not KI.

I haven't done any listening tests yet.

Thanks,
Russ

[AlNewman]

If I had them around, I'd try 6L6 tubes with no dropping resistor.  Judging by your voltages, the HT should be able to handle them no problem.  Your OT, maybe not, but you don't have to test at 11.  Heater voltages may be a thing, but I have an AO-29 here, and it has like 12 tubes in it. 

If you're going to go with 6V6, and buy a new deluxe reverb OT, I would go with a 330R resistor.  If you're buying a 20 watt OT, you may as well get the most bang for your buck, (and weight).  5 watt resistor should be adequate, but I could be wrong.  After that, maybe you could tweak the dropping resistors to the preamp to dial in your sound where you want it.

[sluckey]

CT load = 500Ω          400Ω          300Ω

Pick any of these resistors and call it done.

[CrocMule]

Pick any of these resistors and call it done.

Are you kidding!?  Call it done so soon?  After I install the Deluxe OT, I'm going to do extensive listening tests with various voltages, until I find another project. 

I'm wondering about going in the other direction now, with lower voltages.  How low can I go and still get enough drive at high volume?  I really like the warm throbbing of the lower gain settings on this amp too.  Edge of breakup never sounded so good.

I don't need to post all my opinions about voltage vs tone here though -- I'm sure there's been a lot of that already.

Thanks for your help!
Russ

[PRR]

.......... What's the difference between plate current and cathode current?  I thought current flows from plate to cathode, so aren't they the same measurement?....

Screen grid G2 also passes current. Let's go to the Data Sheet:

[AlNewman]

I'm wondering about going in the other direction now, with lower voltages.  How low can I go and still get enough drive at high volume?  I really like the warm throbbing of the lower gain settings on this amp too.  Edge of breakup never sounded so good.

You could, but you'd probably need a bigger power transformer.  Also, a larger dropping resistor.

[CrocMule]

Stupid mistake!  I was too busy dealing with the field coil PT to think about the field coil OT.  Duh.  The field coil speaker was apparently not 8Ω!  If my calculations are correct, this OT wants to see a 1.6Ω speaker load. 

It didn't click that there are different OTs for the AO-29.  The later ones are for an 8Ω load -- as shown on the schematic I have.  I don't have an early schematic with a field coil speaker, so I didn't know the correct secondary impedance.  To be honest, I didn't even pay attention to that -- my mind was lost in the rest of the circuit/layout/build.

I read somewhere that the yellow wire on "the AO-29 OT" was 4Ω and the green 8Ω -- pretty standard colors -- put it in my notebook.  In my hurry to see if the amp worked, I trusted that note without thinking and hooked up the green wire to the 8Ω speaker. 

I ohmed it out for shorts beforehand, but at no point did I test the OT to verify the winding ratio and secondary impedance until after I pulled it out and installed a new Deluxe OT.  I can't believe I did that.  No wonder the new one sounds different.

My quandary is that the field-coil mismatch OT sounds more like what I want.  The Deluxe OT sounds great at high volume, in-your-face and punchy.  But at lower volume, it's way too stiff.  The mismatched OT is the opposite -- it's overwhelmed at max volume, but warm and purring at lower volumes, and of the sounds I can get out of this amp that's the one I dig the most so far.

With a 69:1 winding ratio, my math says that if the OT was driving an 8Ω speaker load, the power tubes were seeing 40kΩ.  I'm wondering what that does to power tubes.  It's probably not good for them -- probably not a magical combination of great sound and increased longevity.

Does a drastic mismatch like that damage the power tubes?  Could it damage the OT?

Thanks,
Russ

[sluckey]

Stupid mistake!  I was too busy dealing with the field coil PT to think about the field coil OT.  Duh.  The field coil speaker was apparently not 8Ω!  If my calculations are correct, this OT wants to see a 1.6Ω speaker load. 

The OT is 8Ω. See attached...

How did you determine the winding ratio was 69:1?

[CrocMule]

The OT is 8Ω. See attached...

How did you determine the winding ratio was 69:1?

Well looky there -- calls for an 8Ω speaker -- that's interesting. 

I applied 5.90 VAC to the green and black wires of the secondary, and measured 415.8 between the blue and brown of the primary.

With 5.97 VAC on the yellow and black secondary wires, I got 410.0V between primary blue and brown.

Interestingly, I got an open circuit between yellow and green.  The black common conductor is two strands with a coating that needs to be scraped off to get an electrical connection. 

[CrocMule]

After looking at the way the OT secondary leads are drawn on the schematic, I'm guessing (using Uncle Doug's method) this OT wants a 3.2Ω speaker.  Not sure why the discrepancy between the schematic and my calculations.  I used the same method on measurements on the new Hammond 1760H and got 4/8/16 secondaries from 6.6k primary. 

I don't know why the black center tap's 2 solid wires are coated.  Maybe the design calls for the only connection between the two to be made at the solder joint?  Have to scrape them before soldering.  When measuring the secondary voltages, I had to wiggle the clip probe on the CT wires quite a bit to get it to make contact.  Makes sense that I got open circuit between yellow and green -- The 2 CT wires weren't making electrical contact because of the coating -- they were bare only at the cut ends.

[sluckey]

I don't know why the black center tap's 2 solid wires are coated.

That center tapped secondary is actually two identical windings. All four wires are brought outside the bell. The two black wires are the ends of the magnetic wire that is the actual winding wire. The enamel or varnish is the insulation for magnetic wire (same as motor winding wire). The two blacks are joined together to put the two separate windings in series. This creates a single winding with a center tap. Yes, you have to remove the varnish coating from the copper wire to solder the two together. No different from stripping the plastic insulation from the end of regular hookup wire.

[AlNewman]

The field coil only applies to the PT.
The OT is it's own animal.

[CrocMule]

The field coil only applies to the PT.
The OT is it's own animal.

Right.  I guess what I meant to say is it doesn't surprise me that the field coil speaker and OT have a different impedance than what is on the late M3 schematic (without the field coil speaker), since the speaker is a totally different kind of beast.  You are correct, it's just an OT -- nothing field coil about it other than it's driving a field coil speaker.  By "field coil OT" I just meant the OT spec'd for the field coil speaker, to differentiate it from other OTs. 

Okay, I put the AO-29 OT back in (AO-20936-3).  Hooking it up correctly makes a big difference -- imagine that!  I'll leave it in -- gets the warm purr at low volumes and doesn't collapse at high volume.  Perfect for what I want.  I can live with a 2.5x mismatch to a 8Ω speaker.  I might pick up a 4Ω for it.

I have to say, the PT is handy for experimentation.  With the HV CT load, you can change B+ just by changing the value of the resistor.  I tried up to 1kΩ, which brought B+ down to 333V, but it was too soft.  I left the 700Ω in.  With a 5U4GB, it gives a B+ of ~370V.

I measured a max 5W dropping across the CT resistor, but man that 50W sure gets hot to the touch.

Thanks for your help guys!

Best,
Russ