Eico ST70 rework, power supply question

akimball442

New Member
So my old Eico ST-70 which I have had for years, recapped once (years ago) has gotten to the point where it's not worth saving.

When I received it, it had been kept in a garage for about thirty years, and the chassis was rusty. I did the best I could to save it, but at this point, I've decided the chassis is a lost cause and this amp's iron will be better off living on as something else.

So I am also a big fan of Magnavox amplifiers, and have discovered that I can mount the Eico transformers on an old Maggie 142 chassis in such a way that they will be oriented and spaced just like the Eico. Perfect.

Since I like the Eico very much, I'm going to duplicate the power amplifier almost exactly.

But I'm wondering about the power supply. I'll no longer be needing a lot of it, everything for the preamp can go- I suspect I may have to adjust some resistor values to compensate.

Having looked at many other schematics (including many of the Magnavox amps) the power supplies often include a choke. Although I've learned a lot working on these things, and understand the basic function of the choke, I'm wondering if someone could explain to me the advantage or disadvantage of adding one, whether it'd be worth doing or not, and how I would calculate the value?

I look forward to hearing what you guys think.

Schematic borrowed from the internet...
eist70ps.gif
 
A choke maintains better power supply regulation. The choke helps to supply current demands that the amp needs. Many vintage amp used them as it was the tried and true method of power supply design. I think their use fell out of favor during the 60s as mostly a cost cutting measure to remain competitive with the other economy amps.
Disadvantage is mostly finding a place for it. Ideally it should be close to the power trans with the cores aligned at 90 degrees. The value of the choke is determined by current and voltage demands. You would usually add up the current requirement of the tubes and size accordingly. For the eico with capacitor input filter, and only 40uf in stock units, you would probably use a pi filter that is cap, choke, cap. This will give you much lower power supply noise. PSUD2 is handy to show you what voltages you can expect from different capacitor and choke values.
 
In order to use PSUD2 you will need to know the secondary voltage supplied by the PT and also the dc resistance of the high tension windings. You can measure the dcr with your meter and the secondary voltage B+ should be on the schematic.
 
I did what you are considering several years ago for a friend. It's a DIY power amp using Eico ST-70 output iron and the power amp section circuit. We lowered the B+ voltage so he could adapt it to 6L6GB output tubes. It's dead quiet on Klipsch LaScala speakers, which is a bit of a treat given their high sensitivity. It worked out very well for him.
Erik ST-70 top.JPG
 
Thanks for the replies. I will am downloading the PSUD2 software. Sort of funny it's from Duncan's Amp Pages, I've been using the tone stack calculator from there for a very long time with my guitar amp projects.

Quiet is the goal, as I tend to alternate between various highly sensitive horn speakers myself. That looks like a MASSIVE choke.

I'll start a thread as the project starts taking shape.
 
Quiet is the goal, as I tend to alternate between various highly sensitive horn speakers myself. That looks like a MASSIVE choke.

The choke is front left in the pic, so not that large, but the amp is extremely quiet. I think it's because of attention to the signal ground bus and local start grounds.
 
The choke is front left in the pic, so not that large, but the amp is extremely quiet.

The chokes I am familiar with seem tiny by comparison, like little open frame output transformers used in a 3W radio. I have some things to learn.

I'm trying to make heads or tails of the power supply software as we speak.

I think it's because of attention to the signal ground bus and local start grounds.

Could you elaborate on 'local start grounds'?
 
A choke-input filter system would improve regulation but it would also greatly reduce B+ voltage. The original AC line spec was 117VAC, so B+ may be slightly high with no changes, but even if you replace the GZ34 with silicon diodes, a choke-input filter will bring the main B+ line down to something near 330V. One way to make this work is to insert a very low resistance choke downstream of C43A, with an additional big filter cap on its downstream side, and move the OPT center tap connections there as well. Regulation ends up slightly worse, but the amp should have less 120Hz hum and sound a bit cleaner. Another way is to make it a sorta-choke-input, with just a few uF to ground on the upstream side of the choke. You can try tuning this arrangement for correct B+ voltage in PSUD and see how it goes. The first cap's value will be quite critical, so you may be compelled to build it with multiple film caps and fine-tune it after construction. Electrolytic cap tolerance is much too broad for this application.
 
I'd put the choke after the first filter cap, tap the screens and the voltage amp from that point. I'd probably leave the output transformers connected as they are, straight off the rectifier. You'd need a much larger choke if you wanted to run the plate supply through a choke and honestly there isn't vast benefit to doing it anyway. Any resistance between the output transformer and the rectifier is not a great idea, and the big benefit of push-pull output stages is they cancel out power supply noise anyway.

Without the preamp, you can probably just bump the value of R96 and possibly R97 and R95 to set the rest of the voltages where they need to be. A bleeder resistor off point "V" to ground also might not be the worst idea. That would duplicate some of the load of the now missing tubes along with providing a discharge path for the power supply caps.
 
The chokes I am familiar with seem tiny by comparison, like little open frame output transformers used in a 3W radio. I have some things to learn.

I'm trying to make heads or tails of the power supply software as we speak.
Choosing the choke is a matter of adding up your current requirement and choosing the proper voltage limit. Which depends on the power tubes and your operating parameters and the current requirements of your front end tubes. If you employ the CLC filter, the current requirements are not as strenuous as in a choke input filter. Basically, the bigger your choke, or the higher the inductance, the more it will filter the AC. There is much written about this in power supply designs of tube manuals or the "Radio amateurs handbook" and other online sources. Vintage amps built around the 1950s used larger chokes mostly more than 5 Henries and some over 10H. Amps like the Dynaco ST70 have chokes that are 1 to 2 H. If you use a bigger choke in terms of inductance, you need not use so much capacitance to achieve the same level of AC filtering. This makes using and all film cap power supply feasible which i prefer to do.
The psud program is kinda like an easter egg hunt in that you need to put your pointer on various elements and click it to see what are your options. Basically, you input your trans data, type of rectifier and from there your power supply filter. If you are going with pi filter, or CLC, you would input your capacitor (uf). then right click "load", which give choices, click "insert", choose LC filter, then add your choke and 2nd cap data. If you want to stick with the 5K "stock" load hit simulate. I like to use a current load so if you left click you can choose that and input the load for your amp. You can add more RC combos to simulate the input and phase splitter tubes. good luck
 
Last edited:
I think I'm first going to clean up my 'new' chassis, mock up everything, and first build it stock, leaving myself space for the choke in the future. The more I think about it, the more I realize it'd be best to get it working the way it was designed, get it quiet, get it biased properly, etc. If I go through and change things, any potential mistakes will be that much more difficult to spot and sort out.

Thanks for the explanations, they were quite helpful!

A build thread is coming..
 
I think I'm first going to clean up my 'new' chassis, mock up everything, and first build it stock, leaving myself space for the choke in the future. The more I think about it, the more I realize it'd be best to get it working the way it was designed, get it quiet, get it biased properly, etc. If I go through and change things, any potential mistakes will be that much more difficult to spot and sort out.

Thanks for the explanations, they were quite helpful!

A build thread is coming..
You realize that for the EICO replacing a choke with a large dropping resistor is entirely a money saving proposition. Having a choke means better power supply regulation which means better performance and sound. The cheap wire wound resistors used in these amps are also subject to inductance effects since they were simple resistance wire coiled around a ceramic core. As Gadget mentioned, If you just follow the original scheme you will probably have too high a B+ because of the loss of the 4 preamp tubes.
Personally, i would not follow the ST70 schematic mainly because to get 35 watts from a pair of 7591 tubes they had to be run right at maximum limits. This was ok when 7591 were less than 10 bucks but the current crop of 7591 are only a pale imitation of the Westinghouse tube. Maybe if you are planning to use different output tubes like KT66 or 7581 even 6550, the EICO schematic would be interesting.
 
I'm open to any and all suggestions.

But- I can only say what I already know, and that's that the amplifier (when it's not acting up) sounds very good. The problem is that it's just shot, I could change all the tube sockets, probably another handful of resistors, and maybe a cap or two and get it working normally again. But even then, I'll still have an amp with controls that don't track quite right, a corroded chassis, and one that's not very attractive to begin with. So that is the aim of this project.

You realize that for the EICO replacing a choke with a large dropping resistor is entirely a money saving proposition. Having a choke means better power supply regulation which means better performance and sound. The cheap wire wound resistors used in these amps are also subject to inductance effects since they were simple resistance wire coiled around a ceramic core. As Gadget mentioned, If you just follow the original scheme you will probably have too high a B+ because of the loss of the 4 preamp tubes.

Certainly, I understand that it was a money saving measure, and also understand that adding a choke would improve the power supply, hence my original question. I get the impression from the replies in this thread that bigger is essentially better when it comes to the choke, but what I don't understand is how to derive the actual numbers. To be honest, my thought was that if I put it together, I could just measure the current and use that number? Or do I just use the max current specification from the tube datasheets and add them all up? I presume the voltage figure is just B+ (?).

Personally, i would not follow the ST70 schematic mainly because to get 35 watts from a pair of 7591 tubes they had to be run right at maximum limits. This was ok when 7591 were less than 10 bucks but the current crop of 7591 are only a pale imitation of the Westinghouse tube. Maybe if you are planning to use different output tubes like KT66 or 7581 even 6550, the EICO schematic would be interesting.

I understand this- the Eico ST40 uses the same output tubes in class A for 20WPC.

Would I be able to produce more power using some of the other tubes you mentioned, or would I just be operating them within more conservative limits? What other benefits might I realize by changing tube types?

Again, I'm open to any and all suggestions, and very much appreciate the education. I may ask for clarification from time to time, but I'm a pretty quick learner!
 
" But even then, I'll still have an amp with controls that don't track quite right, a corroded chassis, and one that's not very attractive to begin with. So that is the aim of this project. "
The original steel chassis was not plated with a long lasting coating so they tend to rust quickly. But, the transformers are excellent and definitely worth repurposing. They are huge and would be excellent for any 7591, 6l6gc, kt66, fixed bias amp.
Regarding sizing the choke, Yes, you can add all the current requirements and that would be good. In CLC filter the inductor is less "stressed" so that would be more than adequate. A 1 or 2 henry choke would also be ok but a bigger choke would be better.
In regards to power, the operating conditions that you set your tubes will determine the power output. If you are not married to the EICO ST70 schematic then the book is open to many choices. If you like the 6l6 type tubes, the 7581A and KT66 would be a good choice in the EICO circuit if you made changes to the tube socket to properly utilize that power tube. KT66 has .5 amp more heater requirement but i believe the power trans is up to that extra use ( you will have gained some extra current from not having preamp tubes.).
 
Last edited:
Well, if someone asked me what it was, it'd be easier to say it's an Eico on a Maggie chassis! But, no, I'm not married to anything. If there's something you might suggest that would be better, I'd be happy to hear about it.

I did look at the Eico HF89 schematic, curiously, the phase splitter is identical. Supposedly this one is 50W, but it uses EL34s. In the guitar amp world, that's not generally considered the cleanest tube.

I did read about the 7581A you mentioned, and it sounds like it might be an excellent choice. As for KT66 tubes, I have always been curious about them, but I'm afraid the wider bottle might cause things to get a bit crowded on the chassis I intend to use.

If I were to choose to use the 7581A tube, for example, would there be any concerns as far as the impedance of the output transformers that were originally used with the 7591s?
 
WRT the original 35 wpc rating, is that actual or just wishful thinking? Sometimes those old ratings were not all that accurate. Honestly it wouldn't surprise me much to find that you can measure closer to 25 per channel.

The 7591 is easy to drive. Just watch if you change the tube type, the inverter is likely to need to provide more voltage swing in order to drive to full power. The bias supply would also need to be re-worked for mostly the same reason. The 7591 is very high transconductance, meaning large current change at the plate for small voltage change at the grid.
 
WRT the original 35 wpc rating, is that actual or just wishful thinking? Sometimes those old ratings were not all that accurate. Honestly it wouldn't surprise me much to find that you can measure closer to 25 per channel.

You make an important point. I actually obtained this from the original owner, who, to my knowledge built it. He's a former engineer, lifelong amateur radio operator, audiphile, and tinkerer extraordinaire. To think, the poor old Eico was discontinued from use in 1966 when he built a new SOLID STATE Dynaco ST120/PAT4 combo! I had those too, but have since returned them to his son who has taken up interest in the hobby.

Anyway, he told me that the 35WPC rating was somewhat optimistic. He had at some point previously (probably 40 years previously) measured the output and found it to be somewhat less.

There are some interesting articles on this amp out there. I found this website while searching for information, and the author deals specifically with measured power, and how to get it a little closer to the rated output of 35W. Elsewhere on the same website, the author digs into converting the amp to a push-pull-parallel configuration, which vastly increases the output, while still retaining the original iron. He does add a supplementary filament transformer, however. You may or may not find the information interesting.. I did.

....... The 7591 is very high transconductance, meaning large current change at the plate for small voltage change at the grid.

I've seen that word a thousand times, and now understand specifically what it means, thanks for that explanation!
 
I've read the hotrod article with the parallel output tubes. Not totally sure I'd go quite that far, but the power amp tweaks found in the "ST-70A" article would be worth doing. A fair bit of those apply to the preamp circuits, so they won't make any difference to you if you want to make it into a straight up power amp. If you do want to keep it as-is, it would be worth giving it a go.

Never owned an ST-70 of either brand myself but I wouldn't mind playing with an Eico. I sort of have a thing for ugly duckling amplifiers. A couple of my favorites are a pair of modified Bogen mono amps and a nice little Bogen stereo amp. None of them are what anyone would call lookers, but they have it where it counts. I've heard that the Eico's iron is very good, and thats really where it counts. I listen half the time with my eyes closed anyway.
 
Back
Top Bottom