Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Output coupling cap and transformer mod...?
New Posts  All Forums:Forum Nav:

Output coupling cap and transformer mod...? - Page 2

post #16 of 38
2/26/13 at 1:11pm
Thread Starter 

got it... thanks Nikon.

Reply

Head-Fi's Best Sellers

post #17 of 38
2/28/13 at 12:10am
Thread Starter 

I was at my local Frys and they had this Philmore output transformer... for $4.00 !!  I picked it up and thought aah WTH, for $8 I can just toss it if it stinks.

 

http://www.frys.com/product/4785069

 

SHOCKER of the year here.... It sounds really good with my RS1.  The sub-bass (below any fundamental notes) extends a bit deeper and is more resonant as sounds decay through the venue in live recordings.  Everything seems more relaxed too (even for a Grado), the sounds seem to flow more effortlessly with greater depth to just about everything.  I get to use a little more of the volume knob sweep on both my OTL amps, so theres definitely a little attenuation with the transformers.  But its not much at all.

 

Right now I am running off the 70V-10W-500ohm primary, feeding the 8 ohm output to the cans.

 

It has a slew of other primary coils though.  Would I be better off using one of the 25V coils?  Theres a 25V-5W-125ohm... would that one be better for my Earmax amp?  The Earmax is fairly low gain, doesn't have the wattage grunt of the bigger darkvoice, and I think it just runs everything at lower voltages.

 

OK pics... don't laugh at my case-building skills.  I don't have a drill press anymore, and its the best I could do with hand tools.  The transformer mounting tabs slide right into the circuit board slot, and they sit sandwiched between the floor and ceiling with adhesive rubber feet.  I'll get it all painted and mounted up proper soon.

 

Loving this little mod... SOOO much better than in-line resistors as impedance adapters.

 

Thanks to everyone who helped me out... oh yeah the pics.

 

GUTTS

 

 

GLORY


Edited by kramer5150 - 2/28/13 at 12:13am
Reply
post #18 of 38
2/28/13 at 1:02am
Quote:
Originally Posted by kramer5150 View Post

I get to use a little more of the volume knob sweep on both my OTL amps, so theres definitely a little attenuation with the transformers.  But its not much at all.

The attenuation of a transformer is basically the same as the turns ratio, which is the square root of the impedance ratio. There are some losses on top of this though.

So with 500:8 ohm impedance ratio, this will be approx 8:1 which is a fair bit of attenuation (~18dB)  but I do realise you are talking relative to your inline resistor setup that you had before, which has its own attenuation.

 

Going to a 125 ohm primary, same thing, sqrt(125/8) gives a 4:1 turns ratio, so ~12dB attenuation or so.

You should also think about the load on the amp, which is the impedance ratio * headphone impedance, so 500 ohms in this case.

Reply
post #19 of 38
2/28/13 at 11:02am

Nice work! 

 

Glad to see you like it, and got something built for such little $$$. 

 

Quote:
Originally Posted by kramer5150 View Post

It has a slew of other primary coils though.  Would I be better off using one of the 25V coils?  Theres a 25V-5W-125ohm... would that one be better for my Earmax amp?  The Earmax is fairly low gain, doesn't have the wattage grunt of the bigger darkvoice, and I think it just runs everything at lower voltages.

 

 

I would absolutely try the 125ohm tap! The worst that happens is that it sounds worse and you change it back. 

My assumption/bias/first try would be towards the 500ohm tap, but try a bunch. 

 

I did this a long time ago, but connecting the transformer as an autoformer is also fun:

 

 

I'm a bit less convinced as to the necessity of the switch, hindsight being 20/20 and all. Just pick a ratio that sounds good and go with it. 

Reply
post #20 of 38
2/28/13 at 7:26pm
Thread Starter 

Thanks.... Hey "Speco" thats a brand name that goes WAAAY back.  I remember seeing that brand on all kinds of stuff when I was a kid ~30 years ago.  I ran into some of their output transformers just the other day on the WWW.

 

Curious, so the output coupling capacitor serves as a safety net and blocks DC.  Transformers (my understanding of them) also block DC from the speaker.  So if thats the case, can I just remove the output coupling cap from the circuit?... tube output straight into the transformer?

 

thanks

Reply
post #21 of 38
2/28/13 at 7:47pm
Quote:
Originally Posted by kramer5150 View Post

Curious, so the output coupling capacitor serves as a safety net and blocks DC.  Transformers (my understanding of them) also block DC from the speaker.  So if thats the case, can I just remove the output coupling cap from the circuit?... tube output straight into the transformer?

Unfortunately, no. If you allow any real amount of dc to flow through the transformer, it will saturate the core, magnetically speaking, and then be unable to transfer the ac component over to the secondary windings. Apart from that, the dc resistance of the primary winding would completely mess up the biasing of the output tube.

 

If you look at output transformers designed for speakers - those designed for "single ended" operation will be subjected to dc bias currents, and these are much, much beefier construction compared to "push pull" ones.


Edited by DingoSmuggler - 2/28/13 at 8:14pm
Reply
post #22 of 38
3/1/13 at 9:41am
Thread Starter 

OK thanks.

 

 I see this a lot... what does it mean to "saturate the core"?  And what does the output of the secondary sound like when the core is "saturated"?  Is it a form of AC wave clipping?

Reply
post #23 of 38
3/1/13 at 10:17am

My normal round about way of explaining this:

Saturation is a problem because the Core of the transformer can only "hold" so much magnetic field. This field can come from DC or AC. Once you pass that amount weird things happen - its not really clipping, but all sorts of other weird distortions. Perhaps most importantly if you saturate the transformer with DC the AC performance of the transformer WILL be negatively affected. 

 

DC creates a lot of magnetic field in a transformer, so transformers that are not designed to pass DC don't usually deal with it well. Transformers that need to pass DC are designed for it and usually have other compromises in physical size, bandwith, cost, linearity, or some combination of those. 70v speaker transformers are almost never designed to pass DC.  

 

If you are supremely unlucky (mostly if you are using parts near the limits of their ratings or if they are cheap parts and the ratings are "ambitious") you can saturate a transformer with pure AC but this is no where near as common as DC saturation when talking about headphone or "signal level" power&voltage levels. 

 

PS: I happily defer to Doug/JCX/pretty much anyone who understands the physics behind this better than me. 

Reply
post #24 of 38
3/5/13 at 3:27pm

Out of curiosity, have you guys tried swapping in a better volume into these amp? Not sure if there's room in there for a stepped ladder attenuator?

 

KP

Reply
post #25 of 38
3/5/13 at 5:12pm
Thread Starter 
Quote:
Originally Posted by TheKillerPiglet View Post

Out of curiosity, have you guys tried swapping in a better volume into these amp? Not sure if there's room in there for a stepped ladder attenuator?

 

KP

No I haven't.  the step-ats I have seen were all huge though.  I think it would be difficult to fit one in there.  The chassis is not really one continuous volume of space, its top most 1/4 is cut off by a sub-assembly chassis tray that sits inside the main box.   I like the dual pots the way they are set up now though.  I listen to a lot of live recordings and often times one side is mixed hotter than the other, guitar solo way off to the right for example.  Dave Murray solo on the right Adrian Smith solo on the left, Same thing with Glenn Tipton/KK downing, Michael Wilton / Chris Degarmo.  So there are times I fiddle with the volume knobs to balance things out in certain passages.

 

What would be the advantage though... over a quality Alps pot?  My understanding is the biggest improvement from a stepped attenuator is precision matched resistors from right to left.  Doesn't a dual mono design avoid that with the two knobs?


Edited by kramer5150 - 3/5/13 at 5:16pm
Reply
post #26 of 38
3/5/13 at 6:09pm

Ah, I see. I sort of thought there would be little room inside.

 

A ladder attenuator would have just two resistors at any point in the volume path.A non-ladder stepped would have the advantage of using discreet resistors rather than a pot, but they'd be in series. Whether that matters in this case or not is a matter of experimentation. Maybe a pair of Goldpoint Mini V's?

 

At first, I thought the dual volumes might be annoying, but after some use I agree with you. They are handy to have.

 

KP

Reply
post #27 of 38
3/5/13 at 7:25pm
Pretty cool idea. After looking at the Ear schematic I may want to do this for my WA3.
Reply
post #28 of 38
3/5/13 at 9:50pm
Thread Starter 
Quote:
Originally Posted by GrindingThud View Post

Pretty cool idea. After looking at the Ear schematic I may want to do this for my WA3.

Cool... please post back your results if you decide to give it a try.  This little box now has me eye-ing other small OTL amps... always wanted to try an LDII or III.  The only new addition I have made to mine is the addition of a DPDT switch, ON-OFF-ON.  I have the 125 and 500 ohm primary loads wired to the ON positions so I can toggle back and forth different settings.  The Earmax sounds best with the 500 ohm and the darkvoice works best with the 125.  I flip to the middle OFF position when I swap cans.  I have "hot swapped", and even flipped the switch back and forth, neither amp seems to mind... although I don't make doing that a habit.

 

Some other threads of interest.

 

Elliottstudio has some excellent THD and FR plots here:

http://www.head-fi.org/t/629661/a-lo-cost-ht-supply-true-transformer-coupled-tube-headphone-amplifier/90#post_9172063

 

fullcompass.com has a decent selection of affordable speaker output transformers.

 

A mini transformer discussion / build. (I might try one of these just for the heck of it).

http://www.head-fi.org/t/553094/continued-sidetrack-discussion-from-tiniest-portable-amp-i-can-build-nikongod-microtransformer-based-impedance-step-down-box

 

Here's the disturbing thing. I am not really plugged into an amplifier circuit anymore.  I am plugged into a coil listening to an oscillating magnetic field... and it sounds BETTER than the actual circuit I was plugging into before.


Edited by kramer5150 - 3/5/13 at 10:43pm
Reply
post #29 of 38
3/7/13 at 11:30pm

If the transformer increase the output resistance,

 

Is it possible to change the output cap to a lower value and better quality one.

 

I asked because I have a Mundorf M-cap silver oil 4.7uf and Raphaelite output tranformer ( OP5.5K50A,  primary impediance and secondary impedance 5.5k ohm : 32 ohm) sitting on the bench. 

 

eager to try the magic of M-Cap silver oil

 

Sam

Reply
post #30 of 38
3/8/13 at 12:16am
Quote:
Originally Posted by samsie View Post

If the transformer increase the output resistance,

 

Is it possible to change the output cap to a lower value and better quality one.

 

I asked because I have a Mundorf M-cap silver oil 4.7uf and Raphaelite output tranformer ( OP5.5K50A,  primary impediance and secondary impedance 5.5k ohm : 32 ohm) sitting on the bench. 

 

eager to try the magic of M-Cap silver oil

 

Sam

Yes the increase in impedance using the transformer load will lower the cutoff frequency of the high-pass filter formed by the output capacitor. This in turn will enable the use of lower value capacitors, if so desired.

 

The corner frequency (-3dB) point is calculated as

 

                                     f =   1 
                                         2ΠRC

 

 

where f = frequency in Hz
      Π = pi (~3.1415)
      C = capacitance in Farads (so 4.7uf is 4.7x10-6 F)
      R = series resistance in ohms, this will be the load impedance from the transformer + existing output impedance
 
so with a 32 ohm headphone, and that transformer and capacitor, you will have a corner frequency of around about 6Hz, so that should be fine. You wouldn't want it much higher than that though (so don't use an even smaller capacitor).
Reply
New Posts  All Forums:Forum Nav:
  Return Home

Head-Fi's Best Sellers

Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Output coupling cap and transformer mod...?