• 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.

ST-70 build, ideas and advice

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I'm thinking of building a ST-70 type amp. I have a pair of 70's vintage Infinity Quantum 2 speakers I can put in an upstairs bedroom. The room is 20 x 16 feet and I don't need the huge about of volume. So while 35 WPC is on the low side of what Infinity recommended, they were assuming a 70's vintage transistor amp. I should be fine with an ST-70, I think.

My plan is to update the design, use a scratch built aluminium chassis and a beefed up power supply, all the connections on the rear, grounded power cord and so on. But otherwise still a Dynaco ST-70


Here is the question: Which of the common ST-70 upgrade driver broads sounds the best? OK, bad question. What is the difference in sound between the common ST-70 upgrade driver boards?

Are the "clone" A470 OPTs still the best compared to Hammond or Edcor?

Here are two st-70 driver boards I'm looking at
Driver Boards for Dynaco ST70
diytube.com :: View topic - the diytube ST70 driver

Here is a link to the specs on my speakers
Infinity Quantum 2 Speakers
 
frugal-phile™
Joined 2001
Paid Member
Check out Gregg the Geeks ST70, dynamutt, input board. I can't speak for any of the othersm but my ST70 is REALLY good.

As to OPTs, if you can afford them, the O'Netics are very good.

Once you've done everything, the only similarity between an ST70 and the amp you build is that they will be PP EL34 with fixed bias.

dave
 
Triode Electronics' ST70 driver board populated with 2X EF86 and an ECC99 gets my nod. Hafler's topology is retained, while the active devices used are much better.

A pentode I/P stage allows the use of 470 KOhm grid to ground resistors, without worrying about HF losses from CMiller interactions. That, in turn, allows you to use a 50 KOhm passive volume control. As this is for a bedroom system, an "integrated" amp may be your best bet. Somehow, I don't see LPs being a signal source.
 
I use a 12AU7-modified version of the original circuit. I don't need the extra gain because it does not need feedback. I cross-couple the outputs into the cathodes and put a 15 ohm resistor to ground from the 4 ohm tap. So there is feedback at the output (still ultra-linear) but none to the 12AU7 voltage amplifier. Usually with a little cut and splice I can get this to go onto the original board.

The advantage is that there is no loop feedback => sounds better!
 
...

The advantage is that there is no loop feedback => sounds better!

Why would this be? I understand the theory about how feedback should control distortion but I assume there is a side effect?




What about moderate price OPTs? The A470 clones are basically $100 each I consider $100 plus or minus $25 "moderate".

What I want is a nice secondary system. I was thinking of s low power SET and 100+ dB speakers. But then I hate to keep the pair of Infinity Q2 packed away in storage.
 
I use a 12AU7-modified version of the original circuit. I don't need the extra gain because it does not need feedback. I cross-couple the outputs into the cathodes and put a 15 ohm resistor to ground from the 4 ohm tap. So there is feedback at the output (still ultra-linear) but none to the 12AU7 voltage amplifier. Usually with a little cut and splice I can get this to go onto the original board.

The advantage is that there is no loop feedback => sounds better!

Could you please attach the schematic?

Thanks in advance
 
This sounds like yer first build, ChrisA...

I'd stick with what you have to start with in terms of output xfmrs.

My advice is to NOT use the original low level/driver circuit.

which circuit to use is a complex question, and there have been multiple threads on here regarding the ST-70 and similar variants. Read up?

Feedback? Yes it does "flatten" the frequency response, but it causes various other issues in terms of HF transient response. The basic problem is that once the output iron starts to fall off on its own, you can't really correct it properly by boosting the gain in a loop (feedback has this effect).

Of course there is yin/yang in this...

there are also various ways of sending the feedback back, and various places to derive said feedback.

you can go round-and-round-and-round on all of it.

so, the quick answer is build whatcha got.
then do the next one better...

_-_-bear
 
Yes, I've not built a HiFi stereo tube amp yet. But I have some experience with
guitar amps and studied EE in the 70's. I've worked in the areas of radar
and telemetry but not audio.

I can see by looking that the NFB is intended to flatten the frequency response and I suspected it might have some effect, maybe to also compress the dynamics.

I'm trying to understand the trade off.

With guitar amps it's more art than science. I think many times a guitarist will run a tube in a way that according to a load line generates more than 100% distortion so that harmonics dominate.

I have some parts enough to scratch build an ST70 but I don't have two OPTs that match

This sounds like yer first build, ChrisA...

I'd stick with what you have to start with in terms of output xfmrs.

My advice is to NOT use the original low level/driver circuit.

which circuit to use is a complex question, and there have been multiple threads on here regarding the ST-70 and similar variants. Read up?

Feedback? Yes it does "flatten" the frequency response, but it causes various other issues in terms of HF transient response. The basic problem is that once the output iron starts to fall off on its own, you can't really correct it properly by boosting the gain in a loop (feedback has this effect).

Of course there is yin/yang in this...

there are also various ways of sending the feedback back, and various places to derive said feedback.

you can go round-and-round-and-round on all of it.

so, the quick answer is build whatcha got.
then do the next one better...

_-_-bear
 
I built my ST-70 using a 6SN7 driver from Blackburn Audio. For some reason, Blackburn doesn't seem to exist anymore. However, if you want to try this design, I can give you the schematic. I bought some Dynaco transformers from Triode, and I'm very happy with them. I have paired my amp with some 1969 Dynaco A-25s, and they sound great!

I think you'd be very happy if you gave the 6SN7 board a try. I chose to point-to-point the whole thing, and it was rather easy. The only thing I don't like about mine is I chose to have knobs for the front tube's bias. I am afraid someone will come along and "turn up the bass" with one of those knobs and burn out an EL34 and maybe a transformer! Eek!

Pictures are here: Yfrog Album

Kyle
 
The 6SN7 is **way** better than the 12AU7 circuit I was mentioning. You can set it up the same way- 6SN7 gain stage direct-coupled to the balanced phase splitter.

The reason not to use loop feedback is that it exacerbates the odd-ordered harmonics used by the human ear as loudness cues. IOW the amp will sound smoother without it, at no loss of detail. You can't use it to correct for transformer deficiencies, as mentioned earlier, plus the ultra-linear taps and cathode cross-coupling I mentioned do that as good as you need anyway.

If you do something like that I would consider doing the circuit handwired...
 
If you use Dyna O/P trafos, original or cloned, global NFB is necessary. Hafler designed the "iron" that way.

A high gm type as the "concertina" phase splitter is a good idea. High gm provides resistance to slew limiting. With its low gm and high RP, the 'X7 triode is a POOR candidate for the job.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.