Stax Transformer Upgrade

[cjfrbw]

I know I am probably a majority of one on these boards preferring SET amps with Stax transformers over Stax direct amps, however, once in a while somebody asks about transformer upgrades for the standard Stax boxes.
 
I was going to get a Woo Wee.  However, having heard Woo products, I was very uncertain of the transformer quality.
 
Instead, I tried a pair of the new manufacturer Z-565 Dyna OP's from Triode Electronics wired in reverse. I removed the standard stax transformers, and appended the Z-565 trannies to a breadboard for wiring. The Z-565 run about $210 delivered, less than the lundahl and James types often cited.
 
I have to admit, I was always pleased with the standard stax boxes, but these trannies are a significant, not minor, upgrades.  High frequencies are much finer and better defined, soundstage about 20 percent larger, and bass much better defined with lots of internal detail. I have been using it with my fave 300b.
 
For DIY types who also like transformers, it is worth considering. It will probably also require a plexiglass shield to cover the bias circuit, which is exposed in this breadboard.
 
So far I have only been able to try it with standard low bias Lambdas, because the pro socket on my Spritzer mod SRD 7 box has not been working, but plan to get the 507's in action as soon as possible if I can get the pro socket working.

 

[spritzer]

Those look huge which is always good.  The ratio is a bit higher then normal but it shouldn't be an issue.
 
Would you like me to help you troubleshoot the bias supply?  It's the first one I've heard about failing in the field and I'd love to know why.  

 

[cjfrbw]

Hi, Birgir,
 
Yes, the ratio through the 8 ohm tap is 33, rather than the standard stax 25.  The 4 ohm tap would give 66, which is pretty high , but would allow use of 45 tube amp with whatever wonkiness one might expect from a high transformer ratio.  I have used 45 tube amps, and they sounded great with the standard stax transformer.
 
I am only primitive in my DIY skills.  The pro thing started a couple of months ago which seemed like a channel imbalance at first with the 507 headphones.   Any bass notes would cause breakup.  The normal socket always sounded OK with normal bias Lambdas, but would cause some breakup with the 507.  I let the 507  sit a around for a while.  The imbalance part went away, but the pro socket part did not. 
 
I took the headphones to member Shipsut's home, who is now in England, and we tried the 507's with a couple of his pro bias Stax amps.  We both concluded that the 507's were OK, and could not find any problem with them listening at that time, though they seemed to be a bit low in efficiency.  They sounded great, no break up on bass when turned up loud.
 
As you have pointed out, because of the high impedance of the circuit, measuring voltages with standard volt meters is problematic.  However, measuring "leg to leg" in the diode bridge then adding them up, the voltages still were low and added up to about 400. The first "leg" only measured at 85v.
 
I replaced the zeners with 5A 120v types.  With my 120v AC wall socket, that increased the readings a bit, and made the normal bias sound real good. It did not fix the pro bias socket.
 
I have just been replacing parts, which is the only way I know how to approach the problem.  So far, replacing the last four caps and the large cap do not change things.  I am going to try replacing the last four diodes in the bridge today.  If that does not work, then I am back to trying to find a way to make sure the 507s are not faulty, then just don't know.  I would surmise replacing all the major working parts would find the non working part.
 
I am no genius with a soldering iron, so trouble shooting Stax stuff is a bit difficult for me.

 

[spritzer]

Sure sounds like a bias problem and you are on the right track just replacing all the parts.  Since the normal bias is generated off the Pro output via a voltage divider then I'd look at either the ballast resistor next to the Pro output or the cap used to filter it.  Both are rated way above 600V but there are always bad apples which fail prematurely. 
 
120V zeners will give you roughly 700V on the Pro output and increase the normal bias as well to 300V+.  The ladder caps are also running on upper limit of their spec (double the input voltage) so it's recommend backing off to 100V and tweaking the voltage divider instead if you want to breath some life into the normal bias. 
 
I've never tried a 507 with the transformers (at least I don't think so) but my pair didn't go far so I'll try it when I can.  There shouldn't be any issues but Stax change the finer details of the bias setup regularly

 

[cjfrbw]

Thanks, I am gone for the weekend, and will get back to it on Tuesday.

 

[Headdie]

I also intend to upgrade my SRD-6 or SRD-7 box. I'm right at the beginning. Should I understand that the most critical components of an adapter are the transformers ? If so, what are the specifications to look for to upgrade them ?
 
A month later, are you still happy with the Z-565 ?
 
Regards,

 

[cjfrbw]

Quote:

I also intend to upgrade my SRD-6 or SRD-7 box. I'm right at the beginning. Should I understand that the most critical components of an adapter are the transformers ? If so, what are the specifications to look for to upgrade them ?
 
A month later, are you still happy with the Z-565 ?
 
Regards,

Hi, sorry, haven't been following things for a while, I have been gone or pre-occupied otherwise.
 
I think the Z-565 is a significant upgrade to the standard Stax SRD 7 transformers.  They are a lot larger, as pointed out by Spritzer, and that makes them more robust in the bass, but with the amps I have tried, mainly 300b, they just let the great sounding tubes through that much more, I would peg them more transparent to the source. There is lots of more subtle shading with the bass tones than I have never heard before, even compared to high quality dynamic headphones.
 
These are the "new production" Z-565, if you are an antiquarian, some users like the regular old fashioned Z-565 actually used in the dyna kits.  Others say the new production are better, who knows. 
 
Whether they are better than other types like the James transformer types, or the Woo Wee, somebody would have to have a shoot-out, I can only speculate.  I heard the Woo 300b amp with WE tubes, wasn't that impressed.  I have a James 300b amp at home, and I can vouch the James transformers are excellent sounding, but getting pricier.
 
If you don't mind messing with the wires, already have an old SRD with bias board and sockets, and don't mind having a more cumbersome breadboarded unit, it is well worth while.  
 
Still don't have my pro bias sorted out, I am hoping to get back to it next week.
 
In theory, a wide variety of push pull output transformers wired in reverse would "work", but one would have to pay attention to the voltage multiplier and it would be a blind leap until you actually listened to a particular trannie.  So far, the Z-565 is a winner, but impossible to state how it is compared to others. Many of the push pull types designed for 10 to 20 watts would work.  
 
Yes, the transformer quality would be the most important feature of the device.  

 

[Headdie]

Thank you for the quick answer. Hoping you will fix the pro problem.
 
Unfortunately, I don't know much, if at all, about transformers. I'll begin by reading about it.
 
I'll report to this thread before ordering anything.
 
Talk to you later,

 

[cjfrbw]

Well, I have to admit , I am pretty baffled by the "pro socket" dilemma.  
 
I replaced several parts on the pro board.
 
However, the 507's sound OK for a while, and then become less efficient, then start to break up. If I let them sit for a week or so, they sound good when first plugged in.
 
When I listened to the 507's with member shipsut on his pro amp, they sounded fine, but we didn't listen past about ten minutes or so, I noted a bit less efficiency than I was used to when the headphones were OK.  I am thinking maybe I did not listen long enough to allow the problem to emerge.
 
It is annoying to have one pro headset and one pro socket, it is hard to eliminate alternatives.
 
Anyway, after the headphones sat for a couple of weeks, I plugged them into my SRD 7sb normal bias stock Stax transformer box.  They sounded good, no break up, even with normal bias, and played very nicely for a while with good bass.
 
Then I plugged them into my Spritzer mod SRD 7 with the pro socket, they sounded fine, but seemed to lose some efficiency with time.  Eventually, the sound started showing sounds of breakup after about 12-15 minutes or so of listening.
 
Plugging it back into the SRD 7SB normal bias, where it sounded fine at first, it now sounded like serious break up with the same socket that sounded great before.
 
I am now looking at the 507 headphones again, thinking of taking them apart and checking connections or shorts in the bias somehow.  There seems to be some kind of progressive leak or short in the bias wire.
 
Or maybe the stators are failed and only work for a while, but it is hard for me to understand how they would play well for a while then not, both with the normal bias socket and the the pro socket.
 

 

[wink]

Sounds to me like the unit that causes the 507's to lose efficiency and break up doesn't have the 580V bias working.
The headphones work for about 15 minutes on the residual charge in the phones.
Leave your phones alone and check the bias on the unit.

 

[cjfrbw]

Quote:

Sounds to me like the unit that causes the 507's to lose efficiency and break up doesn't have the 580V bias working.
The headphones work for about 15 minutes on the residual charge in the phones.
Leave your phones alone and check the bias on the unit.

That's kind of what I thought, but here again is the experience.  Leave headphones alone for a while.  Plug into two separate devices, a Stax SRD 7SB unmodified unit and then a Spritzer Pro mod SRD 7 unit.
 
When first plugged into the 7SB unit with normal bias, the headphones play fine, with no break up, decent efficiency, and no bass problem.  They will play a while like this.  Then, plug them into the pro socket of the 7 pro mod unit.  Again, they will play fine for a while.  Then, they become less efficient, eventually starting to break up.  Plugging them once more back into the 7SB where they sounded fine, they break up badly.
 
The 7SB unit and the 7 Pro mod unit with a normal bias socket both play fine and sound great with a normal bias Lambda no break up or sound deterioration.
 
That points me toward the headphones having a problem, not the SRD 7 units either SB or Pro mod, and the fact that they play then deteriorate is frustrating, I can't see how that happens.

 

[chinsettawong]

When you say they break up, how exactly do they sound?  Do they sound softer and you have to turn up the volume more and more?  Or is it that you hear some strange noises from the phones?  Are both sides equally loud?
 
Wachara C.

 

[cjfrbw]

Quote:

When you say they break up, how exactly do they sound?  Do they sound softer and you have to turn up the volume more and more?  Or is it that you hear some strange noises from the phones?  Are both sides equally loud?
 
Wachara C.

They no longer sound smooth or coherent, the sound is fractured, starting with the bass and extending to the midrange. Sounds like an old ham radio and not a Stax.

 

[chinsettawong]

How about the loudness?  Is it getting softer the longer you listen to them?  Are both righ and left channels sounding equally loud?
 
Wachara C.

 

[chinsettawong]

I don't think that the problem is with your headphones unless you notice some channel imbalances or you hear funny noises. 
 
Wachara C.