One more Eico ST70 rebuild thread

[gadget73]

OK, grounded cord will sort you then.

The variac won't isolate you from the wall. In stock form these also usually have a cap from one or both sides of the power line to chassis to serve as the "ground". If you don't have that, or if its only on one side and it doesn't happen to be the neutral side, it can tend to be noisier. If you go with a grounded cord you can remove those caps.

 

[oldman55]

I ended up just connecting a separate wire from Eico chassis to center screw of wall outlet, bypassing the variac. Hum went away (though I may have other non-related issues with amp to speaker).

 

[oldman55]

All sounds good but I reach room listening volume when the volume control is set to 2 (out of 10) on some power hungry speakers (Norman 431s). I modified it per the tranola "line stage modification" where I changed out the 12dw7 for a 12ax7. I am guessing that the additional gain in the 12ax7 is causing this?

Maybe I should put a 12dw7 back in or just turn down the input gain or does it matter when it comes to "sound" quality?

One other issue that is not really a problem is that when I set the balance all the way to left channel, right channel plays at about 35% volume. Same occurs when turned all the way to other channel. Wonder if this is also caused by the line stage mod.


Edit: I did not do the "power amp" modification as I thought that all the modifications stood on their own. Maybe not as it include gain reduction.

 

[oldman55]

Anyone do just some of these mods: http://www.tronola.com/ST-70A_mods_package.pdf?

 

[oldman55]

I made all of the tronola mods except for the power amp which I am currently working on. I also did a modification so that the Loudness switch works like one in a conventional amp.
I am thinking of completing it with the exception of adding the AC balance pots (along with swapping out resistors) as I have no way to set them. It's R61a that is highlighted. Question: Will leaving these out affect the other circuits or is it pretty much a stand alone circuit?

 

[gadget73]

You can use the alternative adjustment process for that. It says you put a meter between the test points and adjust for 0 volts.

 

[oldman55]

I believe that is for bias. AC balance would be a different animal.
Since it has test points to set the voltage (and current) and balance pots, I see no need in adjusting the AC balance.
http://www.tronola.com/ST-70A_mods_package.pdf

 

[gadget73]

That pot is for AC balance. it adjusts the plate current and output through the long tail pair inverter. if it was DC balance for the tube bias, it would sit between the grid resistors and the bias supply. Looks like the stock amp already has that in the "N Bal Adj" pot.

 

[RWood]

Gadget is right.

However ........ when I got to the same point as you I ran out of gas before doing that phase inverter mod. I did rewire the OT secondaries, with Common to ground, etc.

I've been using mine without those last little bits, sounds good. The phono is really strong and my main issue before the mods, a lot of white noise from the line amp, is cured. I did end up hard-wiring the tone controls rather than using the PECs. It was a brain-buster but it works very well and now I know it has all new parts with film caps etc.

After going through it all (on both an ST40 a couple years ago and the ST70 this year) my assessment is that they made it much more confusing / prone to problems and errors etc / than they needed to. If I ever do one again I will likely follow BuzzK's lead and strip out most of the extraneous junk and either make a really simple but really good power amp or even an integrated amp but without the idiotic stuff like Balance Check and Center Channel and Phase Reverse. Think of how many people have miswired their speakers because of the way Eico set it up!

 

[gadget73]

I de-cluttered a Stromberg-Carlson integrated that had all that sort of nonsense. It sounded better and I didn't actually lose anything of value. It seems a lot of things from the early stereo era suffered from feature creep. Switches and controls to do all sorts of things that you really do not need. I guess its like bragging about the number of cupholders in a car. You may not need it, but this one has one more than the other one so clearly its better.

 

[nj pheonix]

I got rid of that mess myself on the st70.
Also built discrete pecs, though somebody here did a run of pcbs for them so I have 2 pair of blanks along with EFB pcb spare (Buzzk made).
So in the unlikely event I come across another or a 2080 , I'm prepared.
I'm trying to rationalize it. I like having tone controls. They're pretty much always set flat, but it not there it seems like something is missing .
If a 2080 passed my way at the right price , I'm sure I'd take it. I've been led to believe that Eico fixed what wasn't quite right in the 70 in the 2080.
Really, all the goods are in the 70 to make a better 80 but it'd require significant rewiring ( I fear after I was done and tested and pleased (?) It'd go back on the shelf where it sits now (unused).
A far down the list roundtoit

 

[gadget73]

Tone controls I don't mind if they don't make it sound bad. Some suck, some are fine. My Fishers and Sherwoods keep it, the S-C I had bypassed them at one point but honestly they didn't sound bad so I put them back. The Hoffman console amps sounded awful, but I may revisit that at some point and see if I can fix it.

 

[oldman55]

Finished up on the power amp mods and wanting to try to match up 7591s. But since each pair is balanced, do I need to bypass the balance pot to see the current/power of each tube at a set bias?

 

[nj pheonix]

7591s
In a perfect world they be matched. You don't really have the tools at home to do it. I'd set bias low . Then balance and repeat gradually moving it up to spec. You might want to stay somewhat lower. IIRC .38v sticks in my head . Go, say .35-36

 

[gadget73]

Set the balance pot in the middle of it's travel and set the bias whichever direction runs minimal current through the tubes to start with. If you want to confirm balance pot position, adjust it so the grid voltages are even. Cheater trick, just hang the voltmeter directly across the output tube grid pins. Tweak the balance pot to get zero volts on the meter.

Adjust bias upward to get something close to where it belongs and swap tubes as needed to get them as closely matched as you can without moving the balance pot. Once you've got it as good as it gets, adjust the bias and balance as needed to get it right on.

 

[oldman55]

Set the balance pot in the middle of it's travel and set the bias whichever direction runs minimal current through the tubes to start with. If you want to confirm balance pot position, adjust it so the grid voltages are even. Cheater trick, just hang the voltmeter directly across the output tube grid pins. Tweak the balance pot to get zero volts on the meter.

Adjust bias upward to get something close to where it belongs and swap tubes as needed to get them as closely matched as you can without moving the balance pot. Once you've got it as good as it gets, adjust the bias and balance as needed to get it right on.

I had already put in the 100 bottom plate screws and set it up prior to reading this. I actually put in some questionable power tubes as I figured it didnt matter as far as output. If all tubes are delivering 38ma (setting is .38v and there is a 10 ohm to ground each) at 377v, then they should all be producing about 14 watts. Is it detrimental to the unit running unmatched tubes? The bias on the tubes vary from -15.5 to -16.5.

 

[nj pheonix]

Running unmatched tubes won't harm the amp. Distortion measurements will likely suffer as a result. Whether it's audible, I don't really know.
Fortunately that has the pin jack test points on top. I'm not sure why more manufacturers didn't use them

 

[gadget73]

it'll be OK, just won't measure as well. My Sherwood 5500-II has horribly mismatched tubes in it. I got them to balance out and bias where they belong, but if you put it on the THD meter it reads kind of crappy. Sounds fine though. Stuff that has no means of adjustment can cause you real problems with badly mismatched tubes though.

Weak tubes might be able to bias up to the right current level, but it won't neccesarily mean you'll get proper output power. I've got some tired tubes for a Bogen amp I own. With those in, I get about 70% of it's rated power and the bass is really mushy sounding. Sounds like yours are reasonably OK though. When the bias is in the -12 - -14v range to get them to pass current they're pretty well toast.

 

[oldman55]

Weak tubes might be able to bias up to the right current level, but it won't neccesarily mean you'll get proper output power.

Since math is math, they are all putting out the same wattage (=power). I guess "proper" is the operative word?

 

[gadget73]

Not necessarily. What you're measuring across the cathode resistor is just current. You can multiply that by the plate voltage to get plate dissipation, but that doesn't really have much relation to output power. Thats more a measure of how much current the cathode can provide, which you can sort-of determine by how much grid voltage is required to get the idle current where it belongs. If the grid voltage ends up closer to zero to get the idle current up, it usually means the cathode can't put out enough and the power output you'd measure at the speaker terminals will be low. With a bias and balance setup you can get all 4 tubes idling the same, but once it starts making signal the weaker tubes will stop making power before the strong ones will. That will show up as added distortion if you measure it since one half of the output wave will be stronger than the other. On a scope you'd see either the top or the bottom of the wave going flat at the peak if one tube was considerably weaker than the other in a channel. With reasonably matched tubes, the top and bottom go flat at the same power level.


If you think of it in car terms, say you have an engine that has low compression. It might idle OK but when you put the pedal down its not going to perform like one that was healthy.