Another Eico ST 70 thread

[nj pheonix]

OK a little backround on this one. I bought off craigs list with intention of rehabbing (supposedly working) before it arrived BuzzK seduced me with one of his modded ones which he hasn't finished yet. So I figure i'll just get rid of this one. It sits unopened because of other projects. Bench clears, I open box. What do I find? broken selector knob, broken 7591 other 3 test ? mixed lot, mixed 12ax7s (though all tested good) GE 6SN7s tested good,7247s test good, Japanese rectifier tests good. So I bring up slow with DBT (I put in another 7591) . Voltages seem pretty close to spec. I can't seem to zero with balance pots nor get bias on mark. Is it possible bad 7591s the problem or should I be looking elsewhere. I just want to get it functioning relatively normal and move it to a good home. Any thoughts? Just running into dummy loads now.

 

[shocley]

As usual, I have a ton of questions, but by now some actual experience with this amp.

I also had initial problems with balance not going to zero - I could only get down to 25 mV. My problem cleared up spontaneously, by simply playing the amp for some time. My guess is that the heat generated by those glowing glass things raised chassis temperature and changed some resistance values for the better. Oddly enough, those heavy, bulky black things (a.k.a. power and output transformers) never have gotten very warm.

Now, the questions.

Is this line of reasoning correct? If balance is 25 mV, then bias of the relavant pair of output tubes will differ by 25 mV? So, I can set bias on one to be 360 mV, the other will be 385? And, of course, I can only set bias for the pair of tubes. It looks like I can measure bias for each output tube, though. I could make some measurements, I guess, but I get so infatuated with that obscure feature of the amp called "playing music".

What are the limits of acceptable bias? All Eico says is to set it at 0.38 V, i.e., 380 mV. I know from my experiences with transistor amps that bias has a pretty wide latitude. But I'd still like to know what's what; I don't want to run in class B mode, nor melt the innards of my tubes in pseudo class A.

In testing bias, all you have to do is turn the volume (Level) control all the way down? No worry about any input coming from the program source? I could turn my selector to an unused input, but for me that's a slight PITA, as I also have to turn bass all the way down to avoid an annoying thump. When I switch back, I get a very tinny sound...can't imagine why... :stupid:

 

[nj pheonix]

I don't remember exactly but I think 25mv is close to what i'm getting now . I'll double check. as far as your other question I don't know. Someone will likely chime in. I had one of these, oh about 30 years ago and don't remember having this problem. I'm going to double check some values and see if anything else turns up. Thanks for the reply.

 

[s-petersen]

30 years ago,I didn't know about setting balance and bias in mine.
I couldn't find info on equipment as easily as today,and Sams were too expensive for something that was already "working well",it was my first stereo, $8.00 at the flea market(Stormville NY.)

 

[nj pheonix]

I inherited that one from my father. My local library had photofacts and when I was in Highschool I took a radio and tv repair course. They had a limited photofact library. Otherwise I wouldn't of had a clue (i'm sorry I got rid of that one)

 

[s-petersen]

If not already done,replacing coupling caps might fix the bias issue.

 

[dcgillespie]

As previously mentioned, either coupling caps -- or -- the tubes in each channel are simply mismatched beyond what the balance control can balance out. Its also a good idea to check the value of the 10 ohm cathode resistors, so that what you're measuring is accurate.

Dave

 

[nj pheonix]

I was doing Homework between threads. I discovered 210v was low. a little testing confirmed 120k resistor drifted high. Changed back to value and voltage is right on. Then tested -21 source (both) getting -17.5 on 1 and -13.5 on the other. If i'm reading schematic right, I should have 80vac on xfmr side of diode (which I do) then -110 on dc side. I'm only seeing -47vdc on dc side of diode which leads me to believe maybe 50 mfd cap is bad (shown as c6 on my schematic but not in parts list)(it seems I two schematics on hand with different part #s) (from here on i'm going to go by hifi engine version). Going down that line the 2 68k resistors that go to the bias pots are out of spec .. one is reading 43.3 k other reading 45.5 k so I know they have to change. the 10 ohm bias resistors are all reading 11.8 ohms out of spec but all the same. Not that I have anything on hand to test them with but the coupling caps are .1 400v c19, 20, 21, 22? just for clarification? I'll just pick them up and change them. Do those have to be electrolytics? I think that's enough for one bite...

 

[shocley]

...
Not that I have anything on hand to test them with but the coupling caps are .1 400v c19, 20, 21, 22? just for clarification? I'll just pick them up and change them. Do those have to be electrolytics? I think that's enough for one bite...

I didn't see any polarity markings on those capacitors, listed as "molded" on the Eico parts list, C15-22, and Eico indicates electrolytic capacitors when needed, so I thought they weren't electrolytics. I'd read that the originals were actually some sort of polymer film capacitors, and reliable, but when I asked about it here on AK, the consensus was that they were unreliable and should be replaced.

So, I replaced them with film capacitors, Cornell Dubilier "orange drops". I guess they're the right ones to use. If anyone knows of better replacements, let me know.

 

[nj pheonix]

I decided to replace the cathode resistors and all the electolytics and coupling caps except the chassis cans for now also the old bias supply diode. Waiting on parts ............crap I forgot to order those 2 68ks

 

[nj pheonix]

Just out of curiosity, I was looking at the multi-section cans. I've found direct replacements but, I've also saw higher capacitance cans. So two questions come up, is there any down side to , or practical limit to adding capacity?

 

[s-petersen]

it's my understanding that the first cap after the rectifier should be kept stock,so as not to overload the rectifier,after that increasing capacitance results in less hum,and better stability while under heavy loads

 

[nj pheonix]

Ok first, I was going to move this guy with no more work than absolutely necessary, but i'm becoming attached so I think i'll play with it for a while. (Jon don't get nervous I still want the other one when its finished). So I want to ask, I understand why I have two schematics now. I have the hifi engine manual and the sams photo fact folder. I'm not sure why the components have different names (I guess its not important as long as I follow one or the other. I'm going to try to attach I piece of schematic and ask a question. The diode off the blue xfmr feeding through diode that makes up the bias circuit. It shows 80 vac on line side then -110vdc on load side heading to biasing network. I'm measuring the 80vac correctly. I was only reading -49vdc on the other side of the diode. I'm sure both 68k resistors are out of spec reading 43.3k and 45.5k . Is that enough in itself to read so low (I ordered new resistors and the 3 50uf caps. OK I just answered my own question DOH! the 33ks are reading 22.2k and 23.9k That probably is getting me dragged down. Any comments? I think I'm on the right track

 

[dcgillespie]

NJ -- Unless you take those resistors out of circuit (lift one lead), then you will get a false resistance reading because the two bias control systems are in parallel with each other.

If the AC voltage is correct going to the bias rectifier, and the rectifier itself is good, the most likely cause of low bias voltage is a failed cap on the output of the bias rectifier.

Dave

 

[nj pheonix]

Thanks Dave, Being how I've found so many pieces out of tolerance, I'm just going to go through it and replace anything out. After 60 years all bets are off. I found other resistors out along the path. Just have to wait for pieces to come in.

 

[s-petersen]

if those resistors were lower than expected,your voltage would be higher

 

[nj pheonix]

Thanks Scott, replacing all the big black caps and the electrolytics, also all the associated resistors. Order should come by end of week. In the meantime I'm in a holding pattern.

 

[nj pheonix]

Parts coming in in drips and drabs, waiting for some opinions on further direction, hope to start working on it during the week. Just a brief update

 

[nj pheonix]

This wont make sense but was asking question in PM and couldn't attach images so
I'm adding in questions I originally asked for the benefit of those following :

Hi Dave, I appreciate your answers and continued patience. I think I have most if not all of the near term information. I wanted to get confirmation on what I've concluded. I'm enclosing 2 schematics. (or pieces anyway) one of Jon's power amp section and the same area of of the original schematic (they align nicely so comparing is easy. I'l try to make answering as easy as possible. I'm going to note changes by # If i'm correct you can just say yes.
1 Filter network C35/ R85 showing 120pf v 125 pf Difference probably not critical
2 Filter network added across R53 (8.2K ,39pf)
3 C 33 changed from 25mf to 47mf
4 R 56 changed from 100 ohm to 47 ohm
5 R 61 and 63 changed to 27K with 10K pot to source U
6 Relocated ground connection to C tap of OTF

I understand forgetting the CCS. I am also assuming just from the power amp section that's it and i'm feeding grid of V5 (12AX7) the same.
Please let me know if I got it right . Thanks Andy



(this is an edit). see post 21 for answers

 

[Blackemm]

Did you keep the filtering the same? or did you up it?

IMO, changing the filtering values can change the feel of an amp. My experience (guitar amps) is that lower filtering(even dangerously low) can make for a nuanced amp that sounds smooth and alive, but you may make sacrifices in the the bottom. Higher filtering improves the bottom end, can improve punch but you may make sacrifice in added dimension.

There's a reason we guitar guys play around with our filtering. Although our goals may be different, there is something there, nonetheless.

If you already have it on the bench, couldn't hurt to clip a few caps in and out, to see if you'll change the feel of the amp with different filtering.


Just looked at your schematic. The later schematic added bass, and took off a little highs on that 12ax7. Which one are you doing?
.1 caps aren't electrolytic - but material can affect sound. I wouldn't break the bank, though.
Your leads on your ohm meter can throw off lower measurements. FWIW, my leads/or meter combine for .2 ish Ohms off at 1 Ohm. If yours is more, it may account(in part) for that 10 ohm measurement being off.