I plan on doing alot of testing to get the sound just right before i assemble the entire thing in a chassis but yea definitely safety a very important factor.
Just reading/looking around and the GM70 tube really looks good as the graphite plate version is low cost, and still with a decent output. Also, this tube is the Russian counterpart to the 845 tube, so i could possibly interchange the GM70 tubes with nicer 845's at a later stage and see the effect on the sound.
Not sure if these tubes really fall into the transmitter tube class but they are still large non the less in keeping with the original idea of the thread.
It also helps that there are many 845 shcematics floating around so this will help with design.
Any ideas on how to couple a 845 PP gain stage (class A) with a mosfet ciclotron output stage?
Still would really like to keep this amplifier OTL
thanks again for the help!
Just reading/looking around and the GM70 tube really looks good as the graphite plate version is low cost, and still with a decent output. Also, this tube is the Russian counterpart to the 845 tube, so i could possibly interchange the GM70 tubes with nicer 845's at a later stage and see the effect on the sound.
Not sure if these tubes really fall into the transmitter tube class but they are still large non the less in keeping with the original idea of the thread.
It also helps that there are many 845 shcematics floating around so this will help with design.
Any ideas on how to couple a 845 PP gain stage (class A) with a mosfet ciclotron output stage?
Still would really like to keep this amplifier OTL
thanks again for the help!
never understand wavac or audio note behemoths however can see the point using some exotic output tubes like EF86 IMHO tube sound synergy is more about tube's friendly speakers.
Powering cheap mass market power hungry SolidState speakers with tube amps is the very aukward approach that is maybe why most technically advanced manufactures like cary conrad esoteric manley mcintosh are happy implementing EL84-KT88
Powering cheap mass market power hungry SolidState speakers with tube amps is the very aukward approach that is maybe why most technically advanced manufactures like cary conrad esoteric manley mcintosh are happy implementing EL84-KT88
There are designs for using the 813 at 800 volts PP and SE. see Peter Millets site for SE.
The 845 works very well at 800 volts SE, for 15 watts and only 3.5K .
If you use sensitive speakers you don't need huge Power. Insensitive, sluggish, speakers are best left to Solid State.
Phil
The 845 works very well at 800 volts SE, for 15 watts and only 3.5K .
If you use sensitive speakers you don't need huge Power. Insensitive, sluggish, speakers are best left to Solid State.
Phil
845 is much harder to drive than triode-strapped 813; the 813 has. at 100mA about the same Ra as 845, but double the mu.
Paul
Paul
Hmm any ideas about input arrangement for a 845 PP gain stage?
Would it even be possible to build a 845 input stage that is directly coupled to mosfet output stage (no transformers)??
Would it even be possible to build a 845 input stage that is directly coupled to mosfet output stage (no transformers)??
Why would you want to? 😀
845 super hard to drive, low mu output tube that performs best at obscene plate voltages and has a relatively high rp in addition.. Practically speaking I can't see a good reason, and keeping mosfets alive with the kind of voltages typically used on the plates of an 845 could prove interesting. I could see using a pair as a PP output stage, OTOH I could see saving a couple of hundred bucks and using a pair of GM70 for the same purpose.. 😀
Yep definitely using the GM70s for the design - just thought i'd talk about the design using the 845 as people seem to be more familiar with that tube (even though they're the same).
Just tossing ideas around really at the moment but the idea is to have a hybrid transmitter tube amp of relatively low output (considering the the maximum output of the tubes).
How about a really wacky idea - run the thing backwards so have the PP 845(GM70) output stage being driven by a mosfet input stage? I guess you really couldnt get away with not running an output transformer?
What about a directly coupled circlotron output stage using the GM70s? Maybe possible provided they are driven right?
Just tossing ideas around really at the moment but the idea is to have a hybrid transmitter tube amp of relatively low output (considering the the maximum output of the tubes).
How about a really wacky idea - run the thing backwards so have the PP 845(GM70) output stage being driven by a mosfet input stage? I guess you really couldnt get away with not running an output transformer?
What about a directly coupled circlotron output stage using the GM70s? Maybe possible provided they are driven right?
Member
Joined 2009
Paid Member
That seals the deal - you gotta build it now 😀
output transformer sounds like a good idea to me, with MOSFET driver.
Hmmm it wouldnt be possible to directly couple using a circlotron output stage with the tubes?
You could limit how much voltage the speakers see this way.. although it still would not be ideal.
Any other unusual design ideas?
You could limit how much voltage the speakers see this way.. although it still would not be ideal.
Any other unusual design ideas?
The GM70 is not anything like an 845. 😀 Also its plate resistance is sufficiently high and its current handling sufficiently low that it would not work well in a circlotron.
Why not just build a conventional SE amp with the GM70 - 25W is easily achieved without even going into A2.. Any amplifier based on these tubes is going to be fairly costly to build. Use a variant of George's (tubelab) power drive with something like a C3G or D3A in triode connection and you should be able to achieve > 30W..
All of this presupposes that you are knowledgeable and cautious enough to even (safely) build something around these devices.
I used GM70s in my latest amplifier design, (my first IT based design) and despite nearly 30yrs of experience found out a thing or two..
Why not just build a conventional SE amp with the GM70 - 25W is easily achieved without even going into A2.. Any amplifier based on these tubes is going to be fairly costly to build. Use a variant of George's (tubelab) power drive with something like a C3G or D3A in triode connection and you should be able to achieve > 30W..
All of this presupposes that you are knowledgeable and cautious enough to even (safely) build something around these devices.
I used GM70s in my latest amplifier design, (my first IT based design) and despite nearly 30yrs of experience found out a thing or two..
the 211 has higher mu than the 845, Jolida used a pair to make a 150watt pp amp...
An externally hosted image should be here but it was not working when we last tested it.
Hmm had a look just now and the cheapest 211 tubes on ebay are $99 for a pair which is a bit much... Id rather design using the cheaper (~$20 each) GM70 tubes and later modify the design to sound good with nicer tubes.
Hmm surely if there was enough voltage gain from the mosfet input stage then the low mu of the 845 wouldnt really matter as much?
Hmm surely if there was enough voltage gain from the mosfet input stage then the low mu of the 845 wouldnt really matter as much?
Hard to imagine a nicer sounding tube than the GM70 - nothing at all compelling about current production 211 or 845 compared to these. (They're very nicely made) They're significantly more linear than the 211, and IMLE give up nothing to the 845 - can't imagine why I would design for one and then switch to a very different tube. I would note that Soviet era tubes are very undervalued particularly when there are often no Western analogs. (Exceptions are 6H30, and 6N6 which have gotten expensive) I've learned to dig, and have found a number of interesting devices. (See my Muscovite thread or Wavebourn's 4P1L thread as examples)
In terms of what you propose, the SNR is not going to be stellar cascading a low gain stage followed by a high gain - just the reverse of what you need to do for good SNR.
Oh wow interesting to hear that about the GM70!! Definitely good to hear this considering the great price on them! I was wondering though, is there much difference between the graphite plate and copper plate versions? (apart from the price of course)
And yes not an ideal way of running the gain stages but I am only going for low output (compared to the output of the tubes) so perhaps I can operate them as low gain devices to reduce noise?
Just trying to think outside the box really
Any other suggestions?
And yes not an ideal way of running the gain stages but I am only going for low output (compared to the output of the tubes) so perhaps I can operate them as low gain devices to reduce noise?
Just trying to think outside the box really
Any other suggestions?
I would better prefer ten of 4P1L in parallel, with 600 Ohm 25W output transformer, 250V only B+, than one 845. Cheaper, safer, and more linear. Also, row of identical tubes looks no less sexy than one huge balloon.
Anatoliy
Anatoliy
I should be done with my GM70 build next month and I expect it to put out about twelve wpc. It is not simple build but not overly complicated either. It would be much simpler, but with the heater at 20v and 3a, you can't use hum pots. I am using Rod Colman's DC regulator boards for the heaters.
I am driving the GM70 with a plate loaded 12GM7A and a B+ of 800v give or take.
As for the copper plates, I've read about half and half in favor or no difference, but once mine is up and running for 100 hrs or so, I'll give them a try.
Would your one CF be enough to drive 10 in parrallel? I was thinking 4 in parrallel into Edcor 1.2K/8R 25 watt OT would be interesting.
One source follower will be more than enough to drive 10.
25W OT is huge. However, 15W Edcor transformers are smaller, but have too low primary inductance to satisfy my tastes.
- Status
- Not open for further replies.