• 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.

Dynaco ST-70, Original or Mod?

Member
Joined 2015
Paid Member
Hi All. So today I picked up a couple of amps off CL, rather cheap. A Dynaco ST-70, and a Fisher 660-A. Was just planning on cleaning up, re-capping, new jacks, you know the usual.

First step was to try each out, verify proper operation. The Fisher is all original. Sound? It sounds good, but so do most tube amps that are operating properly.

The Dynaco ST-70 is also completely original, untouched in any way. I hooked it up and I have to say I was really, really impressed.

So maybe I'll actually play with this a little while.

I'd like to hear some opinions on what you all think about sound of the original topology vs some of the others? We all know there are a zillion different driver boards, CCS, triode operation, etc etc...

With most every older amp that passes through the shop, after initial listening it's immediately apparent to me that I can do a lot better. But this ST-70, old caps and surely wandering resistance values, it actually sounds Really Good!

So lets hear some opinions. What did you do, did you like it more or less? Do you like the ST070 at all, etc.

Thanks guys....

Oh, and PS, yes it's my first ST-70. Somehow after building many amps, and owning many more, I've never had one until now??
 
The value ( as for reselling it later) of the st-70 is greatest if left untouched, replacement
of equivalent parts ( powersupply cap , tubes ) .
Sound quality ( as for reproducing natural music ) is most likley best as-is also.
7199 tubes might be a problem, but there still floats around those to a cost that is less
then replacing mainboard , cutting traces to accept alternate tubes will reduce it's value
as st-70 in the same way as any "improvement" does. That said, most 7199 is
in good shape, a DC measurement in place and comparing with documentation will
tell you if 7199 need replacement.
 
Hey Peter. Thanks, I was wondering about the value. I guees I can answer jazbo's question at the same time here. When I say really impressed I don't mean that this thing sounds like one of the best amps I've ever heard. Rather just that I was impressed with how good it did still sound for being all original. The 7199's are fine, I'd assume they've been changed out at some point, probably multiple times even. But the heart of the amp is all original.

I figured there was probably more value in leaving it original, such as with most any antique, electronic or not.

I most likely won't do a darn thing with it. Even re-cap it. After looking last night, doesn't look like there's too much cost return. I'll probably play it for a while just to hear something different, and then let it go.

thanks guys,
 
The Dyna ST70 was never the stuff from which 'the best I ever heard' was made. It has pretty good iron, but a crowded, small chassis, along with other cost cutting measures. It was the lowest cost/entry level way to get a 'high end' stereo tube amp.

After having different mods done to a couple of ST70's, I was never satisfied until I finally decided to build an amp from scratch with the best transformers I could afford. Now I'm sorry I didn't leave my ST70 in the all original state in which I got it, because I think it would be nice to have a period piece like that.

Assuming the ST70 you got is clean and good looking, my advice is to restore the ST70 leaving it as original as possible. If your goal is to put together the best tube amp you can, then maybe you can sell the newly restored ST70 for a lot of money, and use those funds to buy a set of transformers for the tube amp project of your dreams.
 
I've never ben impressed with the stock ST-70. Flabby bass, poor overload characteristics, unique driver tubes, inefficient bias circuit. I did the Allen Kimmel mod (sold by Welborne labs back in the day) and it transformed it into my primary amp for about 10yr.

I kick myself for selling it.
 
Bought an st70 in ~ '71 Used it for years. Eventually Modded it.. Curcio mods, all of them (pricey! ) Thought it was Great, for a few years
In early 90s, brother brought over his just bought NAD integrated.
Completely outclassed the ST70 .. in every possible way. Solid Bass, clear brilliant highs.. a real soundstage for a change.
I shelved my st70 days later.
Technician types at the CBC where I worked then referred to DYNA products as the cheapest possible designs using the fewest/cheapest parts possible.
Being young/arrogant I thought that was just old men babbling.
In truth they were absolutely right.. took me decades to realise that though 😱
YMMV of course.. but imo; unlikely.
 
Last edited:
Yeah, of course it's not the cream of the crop. I've built much better, and listened to much better SS and Bottle. Still though...As was mentioned it does have a few things going for it.

I've always heard about the A-470's being nice transformers. John Atwood's tests showed this to be true. These, and other Dynaco transformers such as the Z-565. I have used them in a few different designs with very good results.

I think it's important to stay relative to what this thing is: A cheaply put together, simply designed, amp for the masses. I know that.

When I fire up any older amp like this, I'm not listening from the standpoint of someone listening to tubes, of hi-fi for the first time. I know the schematic, the sub-par phase splitter, the cheap as it gets (literally) bias scheme, the lack of AC balance, et. all.
for me, I believe that is why it sounded so good. It's not "good period". Like a few of you have said, flabby bass, not the best soundstage to say the least. But when I fired it up, expecting all of this, there was a little magic above and beyond what I normally hear from these old push pull amps. So many of them are very similar topoloigies, I'm sure it's the output transformers, really that's all that it even could be. As we all know, not all iron is made equal. These are some of the nicer examples of old iron through here since the Scott 299 I re-did, as a 6BG6GA amp.

Again, I'm not planning on modding this thing at all. If I was going to do anything I think I'd take these outputs, grab a larger chassis, and go dual mono. I'll keep it for a little while out in the shop, in rotation with a few other bottle and SS amps. Sell it cheap, and still make a few bucks, I got it for next to nothing. But it's fun to hear some of your other opinions.

Who of you has rebuilt around these OPT's? As far as the ST-70 goes, shoot by the time you "mod" everything to be proper, it's not even an ST-70 anymore. But I'd love to hear who of you have taken just the OPT's and designed something from scratch around them.
 
I still have a ST70 that I've stripped down to the sockets, and built a couple of different circuits in it.

One circuit I tried was a PP 6L6GC-triode (cathode biased) with a 12AT7 input LTP and a 6GU7 differential driver stage. I liked it pretty much. The reason it didn't work out was that the ST70 is simply too small to provide adequate cooling without putting the electrolytic capacitors up top somehow.

I've tried a different 'Mullard-style' 6DJ8 voltage amp into 6SN7 LTP, using PP EL34-triode vs. UL, and found that unless I use gNFB, I prefer the outputs triode-wired (which isn't a surprise). The driver had too much gain unless I used it with a 'passive preamp' type of thing (volume pot and selector switch in a box). This version sounded pretty nice, warm and full sounding.

Every time, however, I found something about the sound that bothered me. I think the A470 OPT's are just OK, nothing more. I've looked at them on a scope and saw that they do introduce some ringing on square waves. Open loop, my pair introduce a -3dB at about 35kHz roll-off (if I remember correctly).

I think the next thing I'll try with the ST70 is an 'E-Linear' with UL-tap to output tube grid feedback and 6AC7 pentode LTP.

I was lucky enough to get a pair of Tango XE-45-5 (5k:4/8/16) back in the 1990's. The Tango OPTs are so much better than the Dynaco, it's not funny. Night and day difference. The Tango transformers take the same circuit and make it sound much clearer, less murky. On a scope, square waves look much, much better than through the Dyna OPT's.

I also swapped in a pair of OPT's from a Fisher 800 stereo receiver (I think those are 7k:VC, so not really an apples to apples comparison). The Fisher OPT's sounded better, in my opinion. They ring too, but go up higher in ultrasonic frequencies.

I wonder, would a pair of 25W or 50W Edcor 5k:VC OPT's sound and work better than a pair of Dyna A470's? I think the Edcor's might be better. Has anyone looked at square waves passing through the Edcor PP transformers?
--
 
Last edited:
Huh...I respect your opinion, and I'm not surprised if the Tangos are better...Although not all Tangos are as great as they are rumored to be...True with (most) all OPT manufacturers. When you're engineering a "tradeoff" device...nothing's perfect...

But I highly doubt the A-470 would be beat by Edcors. Subjectively, I have never seen a review stating this, and many that do state the A-470 as being better. Again, it's not top of the line OPT, no. But it is far better than a lot of what was offered back then, and far better than the cheaper offerings of today, in my opinion. I'm sure someone will chime in here...

As far as top end response, 35Khz is already way, way above anything it's ever going to see. I wouldn't want my OPT going any higher for stability issues. That, and my P5 preamp drops off past 20k anyhow...

Personally, I prefer Ultralinear in PP circuits. SE, yes triode. But in PP I like going the UL route and taking GNFB out of the equation.
 
How do you go the UL route and take gNFB out of the equation? I've found that there's not enough damping factor (output Z is too high) using UL operation unless NFB is employed. (I've only tried globally so far.)

Anyway, that 35kHz is -3dB down, so roll-off was about -1dB at 20kHz open loop. It wasn't flat to 20kHz open loop. However, the Tango OPT is. That XE-45-5 model is the best sounding OPT I've tried so far (not that I've tried that many). I should mention that it doesn't do as well as the A470 in the bass. I don't think relatively low powered tube amps are all that great in the bass anyway, so I prefer to maximize their midrange and high frequency sound quality where tubes can really excel. That's where the Tango beats the Dyna.

A friend (whose opinions I trust) built a PP EL84 UL amp using GXPP mid-line Edcor OPT's and he said it turned out surprisingly well. I didn't see the square waves or anything, so I don't know how good 'surprisingly' good is, and how that OPT compares with something like the Dyna Z565.
--
 
Last edited:
At a minimum, the selenium rectifier in the bias supply has to be replaced by a modern silicon part. Selenium rectifiers are ticking, toxic, time bombs. 😱

The 7199 is a problem. They are VERY scarce and costly. The situation with other old stock pentode/triodes is not much better. 🙁 All are out of production. Of the myriad replacement driver boards, my preference is the 1 from Triode Electronics. TE's board retains the OEM signal topology. Populate a TE board with 2X EF86s and an ECC99 (both in production). Better, easily obtained, small signal devices than the 7199 in the original circuit topology is the result.

Scan the archives for the series SS diode tweak. Installing the tweak allows use of low cost Sovtek 5AR4s with complete confidence. OS 5AR4s are not adversely impacted by the "sand".

The OEM C354 choke is a POS. L_RD alone knows how many of those have failed.
 
Global negative feedback is not needed to go UL. Of course UL is a Form of local feedback, albeit dynamic. Yes this does increase output Z. It's a matter of taste I suppose if you're not cranking the amp. Yes, using global will bring down distortion, bring up damping factor (down z), etc. But at a watt or two, when even the peaks in the music are not overdriving the output tubes, I prefer it.

I've heard the same things about the Edcors in a lot of designs. TubeLab talks a lot about this. For the cost (they are cheap), the results are good. I'd bet they are as good or better than many older designs, and obviously than a lot of the stuff today we see in those 200$ brand new E-bay amps from overseas. I was just saying, that from a lot of reviews I have read where guys have played with them AND the A-470, seems the A-470 wins out.

Yes, on frequency response...Any time you can feed an amp only the higher frequencies you are better off. Less intermod distortion. I agree, the distortion signature of Tube amps, especially Triode and UL, lends itself to vocals and higher frequencies. After I started going bi-amped a few years back, I rarely listen any other way. Even out here in the shop, where the Dynaco is hooked up, it's being fed only 80Hz and up (active crossover). Of course, there is still some component of those bass guitar notes contained in the higher frequencies, and this was really where the ST-70 surprised me. I was expecting one thing, and got another. Not perfect, no far from, but pretty darn nice for being all original components. I thought. You hear a lot of that as well on the forums, guys talking about improving the treble response and getting improved bass response....Well yeah, that string being plucked is not purely 46Hz or whatever the note may be...Unless you're listening to purely electronic music with some pure wave component, there will always be some more than one frequency representing any "note"
 
Yeah, that choke is a POS indeed...and does just about nothing more than a comparable R there would do. I always wondered why they even bothered. A lot of old amps were that way though. Was it just that back then the iron for those tiny chokes was cheaper than the larger C, that would have yielded the same ripple results?
 
And yet another opinion... A stock ST70 was enjoyable to listen to, but I could tell it had it's share of Euphonics/subtle distortion in a few ways. I actually researched designed and built what I call a state of the art version of the ST70 from scratch about a decade ago, using 6SN7's and EL34s. It's great, but large, heavy and generates a lot of heat, and has a weak damping factor so the woofer gets to resonate more. The ST70 may be more pleasant to listen to, especially for long periods of time, but in many ways I think my Hafler DH220's are higher fidelity.

If I was given an ST70, I get rid of the 5AR4 rect. tube, and the selenium rectifier (redo the power supply almost from scratch - keep the PT but nothing else), replace all caps, Replace the 2 wire AC cord with a 3 wire, and very probably replace the circuit board with my own design, mostly because of the price and availability of the input pentode. I'd also replace all input and output connectors, and I'd use Banana jacks for the output connectors. Screw terminals on a 50 yr old phenolic substrate sucks. If I used my own circuit for the amp part, I'd also have independent bias adjustments and test points for each output tube. The front end would use metal film resistors and have a passive Rf filter right at the input too. The output terminals would also have an RC shunt (maybe 20 ohms 5W in series with a 0.22uF polypropylene cap), that would limit how high the load resistance could get, as seen by the output transformer, so it would be less likely to ever generate high voltage transients when driven into clipping. The impedance of many speakers goes way up when you get above 20kHZ or so. Transformers get grouchy when they aren't loaded properly. It hurts their feelings, and they get reactive.
 
Why would you buy a 1941 Plymouth?

My brother LOVES his '41 Plymouth. Drives it as much as possible. It is a nice change from boringly identical turd-shape cars from Honyota and Toyonda. The 1930s Dodge and Plymouth WERE very excellent cars for their time. (Ford was still on buggy axles and mechanical brakes. The lowly Plymouth was knee-joint and dual-diameter hydraulics. The '41 Flathead Straight Six runs good as new, which was excellent for the roads of the day.)

Fix the brakes (caps, solders) and drive it for what it is, or sell it on to someone as vintage-minded as my brother. We did used to cut-up ST-70s, like everybody cut-up old Fords into hot-rods, but that is so 1990s now.
 
If I was given an ST70, I get rid of the 5AR4 rect. tube, and the selenium rectifier (redo the power supply almost from scratch - keep the PT but nothing else),

That's what I would do as well. If I was going to keep on this chassis (Again, I wouldn't spend the time) I'd replace all rectification with Solid State.

Eli, you had mentioned something about a diode mod keeping the Bottle rectifier (5AR4). I've seen a few of these mods implemented, diode after the valve rectifier. I guess if you wanted to keep the tube lit up top, but seems to me going this way, I'd just use SS rectification altogether. I know, the slow turn on debate, but I don't believe at these voltages having the voltage come right up is too much harm. I get what you're saying though.

I think I'll just see what the market is like for it, completely un-modded vs. the simple and necessary mods such as the selenium, i/o jacks, and PS caps. Those are all very, very simple things (It's not about complication, rather no reason to spend $ and then instantly sell and not get it back)

It's funny you mentioned the DH-220 Bob. At some point I'll post over in SS about that. I have one here that I re-capped and set to factory spec about a year back, but I've often been curious about some of the modifications for it. It's nice as well, but like the ST-70, there were some cost cutting techniques that need to be addressed if you really want to get what you can out of it, sound wise.

I'm curious what you did with the 6SN7's Bob. I always like to get a peek at a circuit I haven't yet seen 🙂
 
We did used to cut-up ST-70s, like everybody cut-up old Fords into hot-rods, but that is so 1990s now.

Ha! This is the only reason I posted about it and am even taking the time to talk about it. Even though I'm leaving it alone it is fun. When I got into this hobby back in the late 90's, I remember seeing the next modded ST-70, and the next, and the next.....Back then I REALLY wanted one...Instead I got an Eico HF-81, another nice piece for it's day. Bypassed all the crap (there's a lof of crap in an HF-81), and that was the start.....When you start building, and learning, it doesn't take long to realize that with all but a few iconic amps you're really wasting your time trying to turn a Ford into a Cadillac. Much easier to start from scratch and build a Cadillac.
 
The thing about the 5AR4 in the ST70 is that, one of the weak points of that amp is that the power transformer gets real hot. If you dump that tube for solid state rectifiers, you reduce the load on the power tranny by 1.9 amps (the 5AR4 filament). You could leave the un-connected tube in the socket for looks. If grounding is done right, the only difference should be that B+ may be higher than you wanted, so you stick in a power resistor feeding the first power supply electrolytic cap that drops maybe 40 volts (whatever the 5AR4 was dropping with a normal load). Same thing in the bias circuit when you go from selenium to silicon. The voltage drop will be reduced there too.

I recommend a .005uF 3kV ceramic cap (any dielectric) be put across the secondary of the power tranny, so when the diodes switch on and off at the zero crossing of the waveform, any noise they generate will be substantially reduced. I put another of those caps from hot to cold on the AC line coming in, acting as an Rf filter. A 3rd one across the power switch will extend the life of the power switch and get rid of the pop sound when the amp is turned off.