• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Musical Machine (triode PP EL34) question(s)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I am pretty sure that I would like to build a MM with the iron I have. Can someone double check that I'm not way off base? The output iron is has ~4.2K primaries. I know this is a bit lower than optimal. The power transformer should be able to do ~365V if sand rectified...also a bit lower than optimal. It doesn't have to be perfect, just have a reasonable chance of working.

I'm attracted to this amplifier because it seems to be one of the better fits I can find for the iron I have, I have EL34 tubes languishing in a drawer, and its a relatively simple circuit. Also thought about a Cit V. If anyone has any other suggestions I am all ears. I'm not really up to designing something else -- I'm slowly working towards being able to do that but between the day job and kid wrangling I'm best off implementing somebody else's circuit for now.

PS - In case you are wondering, the EL34 version of Poindexter's Musical Machine schematic has been posted in several other threads. No need to re-post it again here.
 
The Cit. 5 is a very nice implementation of Mullard style circuitry. However, 12BY7s are out of production and (IMO) best left for existing units. You can get the same high gm, low CMiller, and large stage gain out of a 6922 in cascode. ;) A LR8 regulator in the cascode's B+ feed takes care of PSRR issues. The 300 mA. of heater current saved can be used to replace the 6CG7 with an ECC99. High gm is a good thing in an amp with a GNFB loop, as it provides resistance to slew limiting from the HF error correction signal.

"Fixed" bias is must, given your low B+ rail voltage, regardless of the topology you settle on.

Send Poinz an EMail and discuss the "Musical Machine" implementation with him.
 
With 4.2K primaries I'd rather use 2a3s or some version of them like the Russian 6B4G equivalents on ebay. 2a3s need humbucking pots to be quiet, but that's no big deal. I also built a similar circuit with PPP 6V6s in triode, like the MM

I've been getting very good results by using the cheap Hammond 124b interstage (half nickel) as a plate choke for a differential pair - just the 90K side which has a centre tap. I've used it with 6SN7 diff pairs so far.

With or without these modifications the MM is a nice design.

andy
 
mfaughn,

I have a breadboarded 6B4G Musical Machine based amp and it sounds very nice indeed. It uses the 6GK5 LTP with CCS in the tail and 5k a-a outputs. I also used grid chokes on the output tubes. My speakers aren't such high sensitivity that I've needed special hum cancelling circuits. I have a single hum null pot in each channel, but it doesn't make much difference. I think separate filament transformers and hum pots for each tube would be required to get it quieter.

Good luck with your project.
John
 
Hey, hey!! The 2A3 / 6BG4 was totally off my radar for some reason. I like that idea. It would be my first foray with a 'real' triode power stage. I'm going to look into that some more. Thanks.

Hey,
I recently converted my PP triode EL34 amp to 6B4G with great success. This was also my first try at DHTs. Input circuit is Allen Wright's PP2C using 6H30. A great choice if you don't mind a bit of "sand" scattered around.
A big advantage of 6B4G is that US NOS is available at very reasonable prices. On my 90dB efficiency speakers hum is a non-issue.
Pete Milletts site has a nice circuit to give you some ideas.
Rod
 
Last edited by a moderator:
Stuffing a few numbers in Broskie's PP Triode software, I get:

400V/60ma/5K (assuming typical MM setup)

21W (seems a bit high)
12W class A
3rd harmonic 3.05%
5th harmonic .132%

365V/70ma/4.2K

16.6W
13W class A
3rd harmonic 2.53%
5th harmonic .132%

How much current can you get out of your power transformer?
 
How much current can you get out of your power transformer?

Hi BW. You might remember a start at an amp design I made a while ago for this same iron. I wanted to use EF86 triode strapped as a voltage amplifier and triode strapped EL34s. You chimed in several times in support of that effort. I'm out of school now and back at work. Less time but no longer augering farther into the red every month. I can afford parts now but a scratch built design isn't in the cards (it was mostly an attempting learning exercise anyway).

Anyways, specs I have on the power supply iron (from Sam's I think):
P: 117V@ 1.8A
S: 140V@ 0.85A AC
S: 6.3V@ 4.4A
S: 6.3V@ 3.8A
S: 48V@ 0.13A DC

Voltage doubler to get ~365V. The amp was donor amp used a quad of 7355 which would have pulled ~300ma I think.
 
I don't suppose you have a schematic for the 6B4G MM?

AFAIK, Poinz has never done a DHT MM.

Anyways, specs I have on the power supply iron (from Sam's I think):
P: 117V@ 1.8A
S: 140V@ 0.85A AC
S: 6.3V@ 4.4A
S: 6.3V@ 3.8A
S: 48V@ 0.13A DC

Voltage doubler to get ~365V. The amp was donor amp used a quad of 7355 which would have pulled ~300ma I think.

No way 300 mA. of voltage doubled B+ from an 850 mA. RMS rectifier winding. Approx. 225 mA. is the limit. Will 225 mA. of B+ be sufficient to feed 4X 6B4s and the small signal circuitry?
 
No way 300 mA. of voltage doubled B+ from an 850 mA. RMS rectifier winding. Approx. 225 mA. is the limit. Will 225 mA. of B+ be sufficient to feed 4X 6B4s and the small signal circuitry?

Eli: Is this a rule-of-thumb? Voltage doubler yields 25% of initial current capability? I haven't had to voltage-double anything yet, just curious......


225ma for a stereo MM would be tough. Would be plenty for a monoblock.
 
AFAIK, Poinz has never done a DHT MM.
I was referring to the breadboard that nerdorama mentioned above. Wondering if he had sketched out his changes or if they were just in his head.


No way 300 mA. of voltage doubled B+ from an 850 mA. RMS rectifier winding. Approx. 225 mA. is the limit. Will 225 mA. of B+ be sufficient to feed 4X 6B4s and the small signal circuitry?
Well, the iron came from an HK TA260, a receiver feeding a quad of 7355 tubes idling at 60ma each (unless I don't understand datasheets at all yet...which is entirely possible) as well as the usual pile of small signal tubes that would be found in a receiver. Could be the Photofact isn't quite right?

This is the main reason I started this post in the first place...to help me figure out what I could and couldn't do with this iron and very hopefully to help me to come closer to figuring it out on my own the next time I have a similar question :)

And also...Thanks Eli. You are always answering my neophyte questions. You are one of the folks that make this forum (and by extension the web) a great thing.
 
Eli: Is this a rule-of-thumb? Voltage doubler yields 25% of initial current capability? I haven't had to voltage-double anything yet, just curious......

Yes, it's a good, conservative, rule of thumb. Remember, TANSTAAFL always applies. When a full wave rectifier feeds a cap. I/P filter, approx. 1/2 the AC RMS capability of the rectifier winding is available as DC, as a consequence of I2R heating considerations. All voltage multipliers are cap. I/P filters and the energy stored has to come from somewhere. I divide the 50% cap. I/P filter basis by the number of multiplication stages. For a doubler, we end up with 25% of the AC RMS current available as DC.

Well, the iron came from an HK TA260, a receiver feeding a quad of 7355 tubes idling at 60ma each (unless I don't understand datasheets at all yet...which is entirely possible) as well as the usual pile of small signal tubes that would be found in a receiver. Could be the Photofact isn't quite right?

This is the main reason I started this post in the first place...to help me figure out what I could and couldn't do with this iron and very hopefully to help me to come closer to figuring it out on my own the next time I have a similar question

I think you misinterpret the 7355 data sheet. Notice that the Class "AB1" anode currents shown are for 2/b] tubes. You can bet H/K went for the power, by using a near Class "B" operating point that resembles the right hand column for "AB1". A 225 mA. B+ rail would have been adequate to power both the 7355s and the small signal circuitry.

With its near Class "A" O/P tube operating point, a MM based on EL34s or 6B4s is pushing your luck, in combination with the power trafo you have. Don't give up the ship! The "Musical Machine" and "El Cheapo" are kissing cousins. Both are a differential splitter/driver and PP "finals". I think we can execute an "El Cheapo Grande", with your power "iron", although 225 mA. is cutting things a bit fine. EC's 12AT7 based splitter/driver works with "12" W. tubes and 7591s. Now, you know where the "Grande" comes from. :D Let's look at the 7591 data sheet to find a reasonable set of operating conditions with approx. 360 V. on the plate. My initial guess is that 40 mA. of cathode current per tube, in either triode or UL mode, is reasonable. PURE fixed bias will be essential, as zero B+ Volts can be spared. 7591 grid to ground resistance limits will force the buffering of the LTP anodes. Otherwise, gain will be insufficient. DC coupled source followers made from little ZVN0545A MOSFETs will do that job very well.
 

Attachments

  • EC big.gif
    EC big.gif
    38.8 KB · Views: 580
Well, that was totally dumb on my part with the 7355 sheet. Pretty obviously says that values are for two tubes. I think I was blinded by wishful thinking.

I am really more interested in using the OPT iron than the power iron. I think the best thing is to just pair it with a new power transformer that is *well* suited for whatever I decide to build. Likewise I'll someday use that power transformer with something that will be happy with what it can provide.

I'm still liking the idea of a PP 6B4G. That pile of EL34's might just stay in the drawer for a long time.
 
The O/P "iron" from the H/K receiver that used 7355s should be OK with PP 6B4Gs. Current production Sovtek 6B4Gs can take a higher plate voltage and a higher plate dissipation, compared to NOS. They go for about $80 a tube. I wish you luck in finding matched pairs of NOS, with all 4 tubes from the same manufacturer and the same internal construction. The really inexpensive way to get close to 6B4Gs is by triode wiring 6AV5Gs. Sylvania actually did that strapping internally to satisfy a U.S. gov't order. No, the triode strapped pentodes will not be as nice as true 6B4Gs, but you can debug your setup without the worry over destroying COSTLY DHTs.

BTW, "Uncle" Ned Carlson has a way to setup an amp, along Dyna ST70 lines, that works with both Sovtek 6B4Gs (no NOS) and EL34s. ;)
 
Hmmm, $320 for a quad of Sovteks or play dice with pricey old tubes...vs. the nice quad of =C= EL34's. Perhaps that won't mean so much after I've dug back out of the hole that going back to school put me in. I have some other horizontal deflection tubes in the drawer -- compactrons. I suppose the possibilities are quite large. Really, I want to start understanding how to build a single stage and how to get different stages to play nice with each other. I know there is a ton of literature and such out there. It is just a matter of picking it up and reading it.
 
Hmmm, $320 for a quad of Sovteks or play dice with pricey old tubes...vs. the nice quad of =C= EL34's. Perhaps that won't mean so much after I've dug back out of the hole that going back to school put me in. I have some other horizontal deflection tubes in the drawer -- compactrons. I suppose the possibilities are quite large. Really, I want to start understanding how to build a single stage and how to get different stages to play nice with each other. I know there is a ton of literature and such out there. It is just a matter of picking it up and reading it.

The O/P "iron" you salvaged from the H/K receiver will be fine in combination with triode wired EL34s. I would, ala "El Cheapo", use a small amount of NFB around the trafos to linearize them and improve damping factor. The 12AT7 based splitter driver does not have sufficient gain to drive both EL34s and the NFB loop. You get around that by using 1 section of a 6SN7 or close relative wired common cathode and don't bypass the cathode bias resistor. Such a setup is a linear stand alone (not inside the NFB loop) voltage amplifier feeding yet another "El Cheapo" variant.

BTW, what sort of preamp do you have or plan on building? The "El Cheapo" variants I've been discussing are "integrated" amps that need only a CDP or phono preamp/TT/cart. as a "2" VRMS signal source.
 
Yea, I had figured out that pentode wired EL34's were not in the cards for these OPTs.

Well, it seems that I consistently find myself really wanting to better understand the whats / whys / and hows of amplifiers. I think I'm once again changing my mind about building 'as drawn' from a tried and true schematic. Having this ostensibly decent but somewhat unusual iron is a driving factor. I could just get some 7355 tubes and build the HK A500 power amp section but probably won't. I'm going to slow down and learn more about how these things work so I can actually know why I would choose a tube for a particular section and whatnot. I've got a DCPP and an RH84 going at the moment anyway :)
 
eBay - The UK's Online Marketplace

I'm getting around $140 from the Siberian Shop on ebay for a quad of 6B4G types. I'm using PP2a3s right now with a differential pair of 4P1L. This is a very detailed sound - your total Russky quality! Got those from the Siberian Shop too. They do need a bit of current, though. Reckon on 15mA each. Mu is around 11, so it needs a preamp. I use a 26 preamp, as described in the "26 pre amp" thread here.

andy
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.