Tube circuit modification

[mitsos]

OK, first, sorry for the ramble, and nobody pull a gun on me for tearing up an old Ampex. I didn't do it. Just trying to help a friend with an already torn up piece of equipment.

Yesterday I was asked to help with a couple of Ampex 601 channels. Just loose modules bought on evilbay ages ago.  I believe one was bought already "modified" for use as a preamp (basically has an IEC socket and had the cable harness that connects it to the tape removed not much else has been done that I can see), the other is stock, from the looks of it. I am against breaking up old equipment as much as anyone else, but this guy is a friend and he's really nice, so, I'm not going to give him a lecture on ethics or historical value. Not yet anyway.

Anyway, the one he's using has a bit of 60Hz hum (way below the signal in actual use, but I think he is wanting to use it on quieter sources).  I checked all the caps and so far they all test within spec (except for one cathode bypass cap that reads completely open). But that valve isn't in the circuit in actual use, it's part of the tape playback.

The thing has AC heaters and they're not wired in any systematic way, so I'll look into that first.


But my gut tells me to clean this thing up by pulling out the unused circuitry (V103, V104, V105, and V107 and related passives) and hardwire it the way that it is being used, move the Rec Cal pot to the front as an output level, and rework the PSU filters to give the specified voltage...  It'll help make room under there and make it easier to see what's going on, besides not wasting those valves for no reason. 


I'd probably swap the sh*tty-looking output transformer for something real-sized (stock one is really tiny, would probably fit inside an OEP case).


One question: Is there any benefit to changing the 5Y3 to solid state? Also, this thing is rated for 5V heaters but is getting 7V.  My thought was to use SS rectifiers, and use this winding with a multiplier for a phantom supply. Am I right in thinking that removing this AC from the DC side of the rectifier would also make the filters more effective, in that there won't be 5-7VAC to filter out?


Soundwise, I've heard this exact channel on bari sax and it was really nice in context.  Having played with it a bit, it wouldn't be my goto pre by any means, but in a larger recording, it definitely can be useful.


thanks to anyone who feels like chiming in.  Just want some confirmation that I'm on the right track here, and if anyone has any experience with these things it's be cool to hear.  I found a couple old threads, but got a warning about posting there cuz they're 2+ years old.

 

[mitsos]

I edited the post for clarity and to remove the part about higher voltages, because I'm not sure that it actually is a problem.  I have 400VDC at the rectifier, 389VDC where it should be 309, and 334VDC where it's marked 260.  But by the time it gets to the 6F5, I have 215V instead of the spec'ed 210.  I have to remeasure V101, I did these measurements yesterday but my computer shut down by itself and I don't have the last few voltages I measured.

Does this sound "normal" or could it be random resistors being off spec after all these years? All old carbon comps.

 

[emrr]

I found a couple old threads, but got a warning about posting there cuz they're 2+ years old.

Maybe I'm hopelessly old school, but I think that warning is really stupid.  It's like a popularity contest or something.  If there's already a thread about 601's, I'm all for keeping it going.  I'm never gonna go find the other threads as a commentator with no vested interest, but will probably scan whatever is in the relevant thread.

I haven't been in one of these in a long time, so commenting on the paper. 

This guy works on a lot of these:

http://fixthatmix.com/Ampex601_DIY.php

I wouldn't change out the rectifier, no real reason.  B+ carried on the rectifier filament works in bazillions of cases.  If you do go SS, B+ goes up, and you also don't have the rectifier filament load on the power trans, will also make everything else go up a little bit.  As far as the other valves, just leave them out and check filament and B+ voltage rise.  If it goes too high, then you can either leave them in, or load the line resistively.  If you go DC you'll have to do that anyway, either way you probably have same heat, different source, either from loading R or higher value series voltage drop R in the B+ line.    I don't know if you'll get that amp 'quiet enough', no matter what you do.  A pentode wired EF86 with no feedback is inherently noisy compared to a reasonable triode, and requires careful tube selection from a handful.  Hum can be from a funky tube as well as poor wiring layout or other suspect parts. 

OK, to benchmark, if you have an outboard PSU (the second 601 unit with all tubes pulled, and PSU caps replaced?), provide external power to the first unit you want to work on, see what happens to noise and hum.  Consider that internal PSU is going to be worse than that, file for consideration.  Maybe try DC filament input and SS B+ rectifier at this point too. 

If gain is far higher than ever needed, and input noise (V101) is too high, consider changing V101 to triode wiring.  Hell, convert it for educational purposes, and convert it back if you don't like the result.  R113/114 looks like a gain staging problem in the proposed usage.  R109 and R113 are voltage dividers inserted to pad down the overall system and also give some isolation between stages; this is a lot of gain along with some purposeful throwaway.  Maybe V102B isn't even needed.  Maybe V102 isn't needed! 

 

[mitsos]

Hey Doug,

thanks for the input. Change of plans due to the protests in sao paulo means I won't get my hands on this again until Sunday night..

But, as for the SS restifiers, and general PSU changes, I live in a country with absolutely the worst AC possible. Maybe India is worse (I've been there, the power lines in the old section of New Delhi are very funny. A rats nest)

So part of the thought behind the SS rectifier is to put less load on the power trafo and post rectification filters (by not feeding them AC directly.  Although, maybe just changing these caps (the 20uF is fairly new, but is brazilian, and not a product I find on the mfg website, so it may be NOS). Either way, I'd rather have a panasonic or something in there. 

Anyway, first thing will be changing these caps to new ones and testing again.  Thanks for the tip on using the other module as a PSU, that's very good idea!

more to come as soon as I get my hands on it!

thanks!

 

[emrr]

Looking at it another way, you are already taking a ton of load off the PT when you pull so many tubes.  I'm personally not worried about the PT load aspect.  If you take too much load off it, you have to burn it with resistors anyway.  I also don't see the rectifier filament AC as being any sort of load challenge for the filtering circuits downwind either. But hey, you can skin this cat a bunch of ways, just don't get too caught up with the problems from one side without considering the problems from the other. 

 

[mitsos]

You're right. I probably am overthinking this thing.  I was just worried that the AC on the rectifier filament necessitated more filtering, and if we removed the AC, the same filters would be more effective on whatever ripple was there from rectification.  But, yeah, I'm only going to do this if it doesn't quiet down after the recap.

 

[emrr]

After caps I find condition of the rectifier tube itself to be the largest problem.  Rectifiers that test well are sometimes garbage in circuit. 

 

[My3gger]

I don't know why would you put so much effort into something that won't work well no matter what you do. This preamps are known for their not so good sound, you found out that yourself.
Both inputs have EF86 pinout, outputs ECC8X pinout, one have OT transformer, one doesn't and both have intestage pot. You wanted to change that transformer anyway and adjust psu.
If this was my project, i would build RCA BA-2 into one channel and EMRR's Gates SA-70/WE141-A mash-up into another. It doesn't take a lot of work beside some trace cutting, psu adjusting, etc. OT for first is pretty cheap and easily available from Cinemag, the other doesn't have it. I never heard Ampex, but i heard the other two and both are very good. They have plenty of gain which is very easy to control with interstage pot and attenuator at the input.
I got 2H chasis from old tape recorder electronics and this is what i will do. Unfortunately both VU meters are missing but i don't care, instead i can put a mesh or something.

 

[mitsos]

I totally understand you, and I wouldn't go out and buy one of these, unless it was really cheap, cuz it looks sort of cool.  But the thing isn't THAT bad. And it isn't mine, it's a friend's, he really does like it but it needs some TLC.  I have really only heard it on one source about 2-3 years ago, in the context of one recording and it was pretty nice. Not so nice now, on my voice with an SM57/58. While I'm no singer, I can tell enough about it from that.  It's not terrible by any means, and it's in really bad condition (only one multicap was ever changed, for another of unknown quality), and as Doug just posted other parts may need checking (rectifier, etc). Another reason to go SS, I can get known good parts for that. If everything is stock, these parts have been through 50 years of use and storage. 

As to why I'm putting this work in, it's part of a trade cuz I need a new interface, the guy has an extra one that I'd like and wants his preamp(s) less noisy. I have to go to the US to get a good deal on a new interface, prices here are 2-3x american prices, due to import tariffs and middlemen.

As for your suggestions, wouldn't you know it, I'm buying the tubes to do a BA-2 soon. See I'm not totally crazy! Except I'm winding my own outputs, maybe that means I AM crazy...  ;D

 

[mjrippe]

FWIW, I have worked on various 600 series Ampex and they can be cool sounding, but not particularly clean.  All of the varieties (600, 601, 602) are a little different and some of the ones I've worked on were modded by others.  The last one I did was made into a mic pre and DI, but only one could be used at a time.  (http://analogaudiorepair.com/2013/03/22/ampex-601-tube-preamp-di-mod/)  The mic pre was from the original mic input (with transformer), the DI was in place of the Line input.  The Monitor Select switch became the Mic/DI switch, the Phones jack is the DI input, etc.  Some new filter and coupling caps were used as needed and "useless" bits removed, but much of the original circuit was left intact.  The result was a decent "colorful" tube preamp which was what the client was after.

Edit: And yes, choosing the best EF86 was important!  The Soviet equivalent tubes (6J32P but sometimes labeled EF86 or 6267) are NOT always good replacements, especially in mic and preamp circuits.  I just pulled one out of a noisy Neumann U67 and replaced it with a NOS Amperex.  The owner was stunned at the difference 8-}

 

[mitsos]

Is that your site? Nice, wouldn't you know, that picture of all the throw away components is what made me decide to remove the unused bit? That and, stock, it's such a mess under there, it's hard to work.  Plus I want it to look nice, I'm a nerd.

And yeah, this guy is not going for clean... at all. just cleaned up, as in noise/hum/etc.

BTW, I assume that you changed it pin 2 hot? These are both Pin3 hot which I've read online is original.

cheers!

 

[mjrippe]

Yes, that's me, thanks!  Removing all the extra crap does give you some room to work in there.  And pin 2 is always hot at my shop ;-)

 

[lassoharp]

And yes, choosing the best EF86 was important

Just using mjrippie's observation as base for a reminder for reference to future readers - Buying a pile of EF86 to cull from at $17 ea or higher - I don't suspect that just because the EF 86 was made by Brimar or Amperex, etc will guarantee it's going to be a quiet one suitable for mic level use.  The brands and era will guarantee a premium asking price though.  I'm pretty sure I've heard CJ comment on EF86 being more noisy than not and having to go through a bunch before finding a good one.  So many sink holes and dead ends.  It would (and used to be at one time) great if you could buy the 600s for 35-75$ ea and get EF-86 for $3-5 ea.  It would still be an amp like the gal you go looking for in the wee hours but sometimes those are good. 

 

[mjrippe]

Yes, I agree that there is no guarantee that a vintage EF86 is a *good* EF86.  My point was actually that the Soviet "equivalent" tube is usually NOT a *good* EF86.  I have had to sort through my meager stash of vintage tubes to find ones that are suitable for mics, preamps, etc.

 

[mitsos]

Yeah. I've seen that about EF86s.. I had thought of the possibility of doing a 2 channel redd47 in one of these, but I have a very limited number of EF86's and I don't know how the ones currently in the units will end up performing once cleaned up.  I'm curious to hear the difference myself. One of these will stay "stock" while the other will probably get a completely new circuit.  I got overwhelmed with work these days, and have some work travel this week, so I won't get back to it until later next week, but that gives me time to get the parts and think about the best route for these.

Lassoharp, I sent you a PM (or was it an email?) re: the Altec 458 paper, did you get it?

 

[lassoharp]

Hi Mitsos,  Yes I got the 458 paper some time back.  Very good reading.




6au6 are plentiful and decently priced. 

 

[mitsos]

I think I just might have some of those.  

Have you used them in anything similar?

 

[lassoharp]

Haven't tried them yet.

 

[mitsos]

short update... got home after more than a week away, stayed up late and finished this thing. Ended up removing all the "unused" parts and setting it up with just the 3 tubes. V101, V102, V106. 

HT filter is now CRCLC (added a 10K before the existing CLC to bring the voltage down a bit).  Also running non-regulated DC heaters now.. 60Hz hum is gone, I have a bit of 120 at high gains and some noise that is probably wiring/grounding related.

I removed the line input and pot and took R110 straight to ground.

Swapped the 250K "rec Cal" pot for a 100K with an additional 100K resistor to ground. So with the 100K prior to the pot, there is 300K total to ground. I basically centered the pot's range around the fixed setting it was at before, so I can drive the output a bit more or less than before. We'll see what eventually makes it to this spot in the front panel.


It's a bit noisy, probably have to do with grounding/wiring.. Should make it a bit neater. The thing was grounded to the case everywhere, I mostly kept that going but I wonder if running a ground bar or taking each stage back to the main filter cap ground might be better? (never mind, had a problem with mains where I was working... all fixed now).


Also, I wonder if the cathode R and bypass C for V102A should be changed? It was shared with V103A, this latter tube is no longer in circuit. I tried raising the resistance, didn't work too well. Too much gain IIRC. Should take better notes.

cheers! Any hints welcome! 

 

[mitsos]

Another update now that it's been in use for a while. Noise mentioned above was bad mains wiring at my place, finally got it fixed a few weeks ago and now I can breathe easier.

The 100K pot + 100K resistor remained in place of the 250K rec-cal pot. Works very nicely.  My buddy absolutely loves it, but there is one issue.

In an attempt to lower the hiss at higher gains, I attempted to lower the gain of the first stage (EF86) by messing with the R105/R103 voltage divider. IIRC I raised R103 to 100K... This lowered gain and removed the hiss, BUT it also lowered headroom below ideal. The thing sounds great if you  keep it in its sweet spot, but I wonder if maybe feedback would be a better idea here? Or changing R104?