Jbau: your results are consistent with ringing at about 10 kHz. My findings were similar. The fact that 300 pF vs. 100 pF termination doesn't substantially change ringing frequency indicates that transformer's self-capacitance dominates.
And I believe we earlier agreed that SRD transformers are of poor quality.
And I believe we earlier agreed that SRD transformers are of poor quality.
@jbau - thanks for posting this and the write up at your blog.
Can I ask - the RC network you recommend on the secondary of the transformer at the socket with the 610k-1000pF - what is the actual voltage likely there? Would I get away with 100V fkp2?
headphones are the original lambdas
Can I ask - the RC network you recommend on the secondary of the transformer at the socket with the 610k-1000pF - what is the actual voltage likely there? Would I get away with 100V fkp2?
headphones are the original lambdas
Hmmm, my hunch is a 100V cap won't do there. You could easily see 150V RMS or more in loud listening, though it may be less across the cap with the series R. If you have the manual for your Lambdas, they probably give the sensitivity spec, as dB SPL per 100V RMS. The SR-X is 97dB.
Last edited:
The response peak is at 15-16kHz. Not substantial enuf?
With the difference that I see these transformers as quite workable and with proper optimization, give good results. Good luck in your search for a superior 50:1 voltage stepup.
I had good luck with UTC A-12. This is an input step-up transformer with center-tapped secondary. Hammond 806A and 808A are other examples of this kind of transformer. These are all good quality pro-audio units with high permeability cores.
Another option is a good quality push-pull output transformer designed for low power tubes such as 6V6 or EL84. An example could be Lundahl LL2766.
However, I did not pursue this path. A push-pull triode amplifier produces balanced audio voltages needed to drive STAX phones directly, with no need for step-up transformer. Moreover, output tube DC voltage can be used to energize the membrane, with 1M resistor to ground.
What I meant was that ringing frequency can be deduced from 1,000 Hz square wave form. Ringing is seen as decaying undulations at the crests and troughs. One full period of ringing spans about 1/5 of square half-period, so the ringing frequency is about 10 times the square wave frequency, or 10,000 Hz.
To further dispel the notion that these transformers are so poor as to be unworkable, here's harmonic distortion of the modded SRD-7 at 1kHz and 100V RMS output. A good result, I'd say. Going by the amp's volume pot setting, this is the level that I normally listen at, corresponding to ~ 99dB spl with the SR-X cans.
Attachments
Another update: A more accurate way to correct for the tonal imbalances of the SRD-7 / SR-X MkIII combo is with digital EQ, such as the excellent (and free) Equalizer APO ( https://sourceforge.net/p/equalizerapo/wiki/Documentation/ ).
Configuration and filter files for the EQ have been added to the instructions. This is the best-sounding option by far...
Configuration and filter files for the EQ have been added to the instructions. This is the best-sounding option by far...
Yet another update: Mistakes happen. I discovered post facto that I used the wrong calibration file for the network analzyer when making one of the measurements. I redid the measurement. Fortunately, only two components in the optional passive filter circuit changed value. Everything has been updated.
I had a look in my SRD4 for comparison. All that was in circuit before the transformers was two thermistors (red discs, so I presume) in parallel for each channel. I have replaced them with a Kiwame 5W 4.7R each which seem to work well so far.
Interestingly, my SRD6 is also missing the 27R resistors as well.
Thanks for the right up on your mods jbau. I have a set of new 50:1 very high bandwidth transformers Jack from Electra-Print made for me over a decade ago... Just the two transformers are larger than the entire SRD6 so I need to find a new enclosure for them. I'm going to transfer the bias board over after I do the mods and I'm going to remove the speakers selector switch and speaker outputs as well. I just need the energizer.
Thanks for the right up on your mods jbau. I have a set of new 50:1 very high bandwidth transformers Jack from Electra-Print made for me over a decade ago... Just the two transformers are larger than the entire SRD6 so I need to find a new enclosure for them. I'm going to transfer the bias board over after I do the mods and I'm going to remove the speakers selector switch and speaker outputs as well. I just need the energizer.
That's interesting, BMF. Not easy to get wide bandwidth in a 50:1, takes special winding techniques which are not usually used with high turns ratios.. Would like to hear about the results. Measurements would be great if you can.
It might be a higher ratio since they were designed to pair with my 35W/ch Yamaha CR-620 receiver or the Pass F5. My SRD6 is marked for a maximum input voltage from the amp of 10V, so the 50:1 makes perfect sense there.
It's been far too long, but I know I spoke to Jack about it several times and it took him a couple attempts to get there. He understood what I needed and the application. Most importantly, he was up to the challenge. I know they weren't cheap and I have no idea what he did to get the bandwidth I wanted, but he sent me the results of his testing. I wish I could remember the spec for the desired bandwidth in KHz. I know I had help from a couple folks determining what the ideal minimum bandwidth should be. I'll see if I have any records in my email...
Now I just need to listen to my SR-Lamdas through the stock SRD6, then swap out the transformers and listen again. I'll just mount them to a board next to the SRD6 for now. 😉 After that, I'll do the circuit mods.
I have no way to measure the final results... just my ears and my perceptions.
One other thing I'll mention is that I have an AT-706 energizer for electrets and an old Teledyne energizer for electrets. Both have significantly larger transformers than the Stax. The Teledyne doesn't have a center tap on the secondary. The AT-706 has 3 leads on both the primary and secondary. I can use a DMM to see if one or both are CT'ed. If only because now I'm curious. I also have a variac so I can safely see what the ratios are for the three sets of transformers. I just have to remember where I stashed it.
It's been far too long, but I know I spoke to Jack about it several times and it took him a couple attempts to get there. He understood what I needed and the application. Most importantly, he was up to the challenge. I know they weren't cheap and I have no idea what he did to get the bandwidth I wanted, but he sent me the results of his testing. I wish I could remember the spec for the desired bandwidth in KHz. I know I had help from a couple folks determining what the ideal minimum bandwidth should be. I'll see if I have any records in my email...
Now I just need to listen to my SR-Lamdas through the stock SRD6, then swap out the transformers and listen again. I'll just mount them to a board next to the SRD6 for now. 😉 After that, I'll do the circuit mods.
I have no way to measure the final results... just my ears and my perceptions.
One other thing I'll mention is that I have an AT-706 energizer for electrets and an old Teledyne energizer for electrets. Both have significantly larger transformers than the Stax. The Teledyne doesn't have a center tap on the secondary. The AT-706 has 3 leads on both the primary and secondary. I can use a DMM to see if one or both are CT'ed. If only because now I'm curious. I also have a variac so I can safely see what the ratios are for the three sets of transformers. I just have to remember where I stashed it.
Hey jbau,
I went back and searched my email. Sorry, I don't have any of the emails about the E-P transformers... I'll try to find them and see if I can do any simple measurements on them. The transformers I had handy were a different set that Jack made for me for another project. Fortunately I marked them because they were very similar in appearance to the step up transformers for the Stax energizer.
We moved about a year and a half ago so I've been trying to find all my diy goodies for the last 4 weeks or so in order to start building some stuff again after an 11-12 year hiatus from the hobby. During the move I purged a bunch of stuff and condensed most of it into 3 large totes and 3 small totes.
I went back and searched my email. Sorry, I don't have any of the emails about the E-P transformers... I'll try to find them and see if I can do any simple measurements on them. The transformers I had handy were a different set that Jack made for me for another project. Fortunately I marked them because they were very similar in appearance to the step up transformers for the Stax energizer.
We moved about a year and a half ago so I've been trying to find all my diy goodies for the last 4 weeks or so in order to start building some stuff again after an 11-12 year hiatus from the hobby. During the move I purged a bunch of stuff and condensed most of it into 3 large totes and 3 small totes.
Do I understand correctly - the best custom OPT instead of original one should be:
1) un-gapped
2) has winding e.g. 4 Ohm:10K(CT)
3) be 10+Watts?
1) un-gapped
2) has winding e.g. 4 Ohm:10K(CT)
3) be 10+Watts?
This is what I got when I measure the Input impedance of a STAX SRD-7 pro mk2, doesn't seem to match to what was posted here, hmm
Asssuming your measurement setup is correct, was the transformer output terminated, with headphones or 75-100pF ?
Check both channels to see if they are the same.
I'd check the impedance of the PTH. From what I've seen, they go high with age.
Check both channels to see if they are the same.
I'd check the impedance of the PTH. From what I've seen, they go high with age.