GM70 Monoblocks’

[Paul Barker]

quick image upload

So discovering how much power is available in A1 SE and how low 2nd harmonic at the Maximum Power. The project worth the effort. Many years ago I bought the power transformers pictured, likewise the Peter Tribute OPT’s. Actually the room Im sat listening to the 6550 amp on the Quads is where
Peter slept when he delivered them intime for Eggborough when we both attended and I brought a PX25 amp resistorless capacitorless after Stephie Bench, and a V1505 amp. On a version of Ariels that were PANTS.

[Nick]

38W @ 0.03% THD, £5 + Heatsink and power supply.

Just saying

[Paul Barker]

The first difficulty is a matched pair of power chokes capable of choke input. If anyone has bought such a pair and regrets it. paulv1505@ insert the applemail ending here. I have a pair of cores in a double C core transformer. One core of that monster per channel would make said pair of chokes. But someone here may have a pair and I’d rather not have to wind them.

The power transformers are rated for 250mA each. The output stage probably as per drawing Quiescent 1,200v 100mA.

I don’t see the need to extend the amplifier into class A2. These power transformers raise a GM70 design into a goodly power out with a good distortion expectation. Certainly the 6550 Im sat glowing while I listen to my reward for all the hard work it cost me. But it now rewards me and all the catastrophies and mistakes are repudiated. As my BJJ Teacher son Sam says, “you learn as much when you don’t win a match.” Also a saying Owen Livsey in Hull tells his guys when they compete.

I fought off all the failures of the 6550 fiasco right up to the mistake of not timing the HT automatically. With help from my friends here.

So shant have to drive grid current.

This amp requires greater voltage gain, but I can see at least three stages. Not having done the math yet, I may start with Mark’s favourite the 4P1L dh pentode input valve, follower by a dht maybe 801a with its mu of 8 driving the output valve. But havent done the maths on sensitivity yet. Its early doors.

[Paul Barker]

38W @ 0.03% THD, £5 + Heatsink and power supply.

Just saying

Brilliant and a lot cheaper to build I guess.

[Paul Barker]

So Ive drawn a 4k resistive load. +150v anode and g2 -6v bias. Not having enquired anywhere as of yet. But 2v peak seems to convert to 150v peak. GM70 needs 120 volts peak. Assuming the 10y mu of 8 I can get gain of 6 from. Not having calculated the circuit yet. In hope that the 10y should counter the gm70 second. The pentode provide hoped for timbre.

[Paul Barker]

[Paul Barker]

Ive reworked out the. Distortions. 2nd is correct. 3rd is lower, 0.2%. I wont edit the post above as the thrust is, that in Class A distortion is not an issue.

[Paul Barker]

Updated

spacebar and backspace not working

[Paul Barker]

[Paul Barker]

I’m thinking about the driver, and looked at the 45 the px4 aswel as the aforementioned in my first sketch. But I’ve now decided to use the EL34 triode connected. Since the 400v b2 supply coming straight in at that position is perfect for it. It gives the voltage swing I need for full power of the GM70. I thought long about the 6EM7 stage II but the b2 at its beginning being 400v wouldnt suit it, I suppose I could have dropped it to 150v with resistor loading, but couldnt get a pentode ccs out of my brain, after a long night awake I thought “kiss”, better to make it at all that get too trixy for my own good. That’s when I was reminded just how much I enjoyed the EL34 28 years ago on first hearing a valve amp at my friends (Richard Warwick who made an unusual plasma tweeter, timeteck on Youtube) house he just hooked up a single valve which happened to be whatever he found rumaging and it was an EL34 and Ive was hooked. Of course it was Mullard then but now I have only Russian version, and only four.

Anyway circuit same, driver valve changed. No worries about filament hum, leaving only the GM70 as a dht, which will be heated by laptop supply right away.

The power supply for the first three stages commences at 400v, well clearly the b+ for the EL34 amp (started life as 6550 which didn’t sound right, too clinical but stronger) so that is a seperate supply with Black Gates any voltage I want up to 630 choke input with a very bold and hearty doube c core choke of large dose of Henrys’ which silently regulates the downstream world, and a 500mA transformer. Just plug it into the voltage amp driver chassis, which shall be made more beautifully on a single use other Nightingale chassis, as yet unused, with the ae interstage transformers mounted instead of output transformers. Shall come out neater than the version of that chassis which I used as a breadboard to build the EL34 on eventually, and that is a keeper for every day.

So there is a vacancy for a hearty output chassis which houses the Tribute monster OPT’s and the valves.

Then another power supply for that output stage. The pair of 2920 CT shall work together phase one of the build with U19 rectifies and a single 500mA potted military choke. 1500v Polypropylene caps for the smoothing Pre69 alerted us to in here many years ago.

All this because I dont have a pair of suitable chokes right now and that monster will nail it.

This is certainly looking heavy and heavy duty already.

Should be just enough voltage to autobias, may require a balancing act with a few microwave caps to lift the b+.

Anyone know if the copper gm70’s can dissipate 125 watts?

[Nick]

I think they can, though, I would think you might want to run that o/p stage b+ via a variac partly for the valve, and partly for room temperature. 125W in winter, maybe a bit less in summer (+ 60W heater power as well). In the amps I built Colin way back, that Andrew has now, I used a similar circuit, except I used a 300b as the driver and a c3g to drive the 300b, as was fashion at the time, If I did it again, I might do what the 211 does and use a dual rail front end and direct couple from a cathode follower. Then again a 211 needs less swing for full power.

[Paul Barker]

Yes the 300b was considered also and the px25. But EL34 is ok at this phase. Then since theres a interstage I considered Sukuma style with GM70 . But this should work.

I’ll put the big blue variac we used with the gm70 we tortured at eggborough on Colins inefficient electrostatics. These fifty sevens are much more lifelike. We also used that on the 833. That huge 3000 v transformer is still at the lockup.

No need. But I cant bring myself to part with it.

[Nick]

I would hazard to suggest the el34 would be better than a DHT, far less microphonic as a driver valve. Why not just do what Phil does, make your existing amp the driver, replace the output TX with the interstage, then just add the gm70 output stage. At least as a prototype.

[Paul Barker]

I think it needs the extra gain of the 6072 but I could leave it as it is and get some volume. Im glad you agree about the EL34 and that you and I are on the same page. I agree a quick first pass is the way to go for me in my state.id like to try and get the tributes and output valves done tidy if I can. I could build up the driver chassis one stage at a time. Start with what it is add the interstage but I think it wont be difficult to make a four pin chassis socket to quickly swap from output transformers to interstage.

First alteration to signal section 6072 for more gain.

Meanwhile explore chassis for GM70 heater and output transformers.

Meanwhile enjoying the EL34.

What put me off the obvious 300b was it isn’t exciting me. I’m half regretting casting off the 801a but its the hastle of so many dht hum to avoid as well as I can. That and the EL 34 is doing its job and sounding great. Very quiet as it is and good rythm and all that stuff.

[Paul Barker]

My state of mind is much better now I’ve weeded out the failed components and arrived at something that sounds better than the pse 801@ and that Ive put to bed class A2. These GM70’s do such a good job in A1 I see a good amp at the end of the tunnel.