[bobkatz]

Next, THD (without noise) at 1 kHz. First a reference THD at 1 kHz at 100 dB SPL. Note that I use 100 V RMS reference = nominal 100 dB SPL for my Y axis. I think that expressing levels and noise in equivalent SPL for the EStat headphone makes a lot of sense. THD is 0.003% at that level at 1kHz, which is exemplary, especially for a transformer-based amp. Then I push the level until I reach 1% THD and back off to just below that. For the 1 kHz, that's a fantastic 119 dB SPL -- excellent headroom, THD at 119 dB SPL is 0.03%. This is just for momentary transients, we never play even close to that loud.

Next, THD (without noise) at 20 Hz. This is really pushing this old transformer. The max SPL I can get at 20 Hz below 1% THD is at 108 dB SPL measuring 0.6/0.5% THD. The harmonics look really ugly at 20 Hz at this level. This is clearly a weakness of the amp, if you consider that pushing above 100 dB SPL at 20 Hz is a goal (not mine). At a more reasonable 100 dB SPL, at 20 Hz, THD measures 0.03/0.04%, consisting largely of 2nd Harmonic 90 dB down and 3rd Harmonic 70 dB down (clearly the saturation of the transformer). There are other ugly harmonics shown in this picture which I'm sure affect the tone somewhat.

 

[bobkatz]

Next, THD (without noise) at 20 Hz. This is really pushing this old transformer. The max SPL I can get at 20 Hz below 1% THD is at 108 dB SPL measuring 0.6/0.5% THD. The harmonics look really ugly at 20 Hz at this level. This is clearly a weakness of the amp, if you consider that pushing above 100 dB SPL at 20 Hz is a goal (not mine). At a more reasonable 100 dB SPL, at 20 Hz, THD measures 0.03/0.04%, consisting largely of 2nd Harmonic 90 dB down and 3rd Harmonic 70 dB down (clearly the saturation of the transformer). There are other ugly harmonics shown in this picture which I'm sure affect the tone somewhat.

Moving to a much safer frequency, 50 Hz, performance is instantly much better, and considering there aren't many 20 Hz fundamentals in typical music, especially at such loud levels, I think the performance at 50 Hz is a more realistic demonstration of what to expect from a well-designed transformer-based amp. As expected, the higher we push, the uglier the harmonics look on the FFT. A safe indication of its headroom is, I think, 110 dB SPL at 50 Hz, with 0.02% THD. You can push it to 116 dB SPL before it reaches 1% THD, with 0.29% THD, but it looks really ugly in the FFT. Let's hope we don't need that much low frequency headroom! Listening tests to all kinds of music are sounding just fine, with good, clean bass definition.

 

[bobkatz]

Thanks for sharing results, been curious about toroids after someone mentioned it in one of my videos, haven’t really done square wave testing at anything other than 1khz on mine but there is a little overshoot that settles quickly on 1630s. That said if you’re using a DAC etc it could just be amplified from the signal chain also.

I tend to run tests at 150Vrms (seems to reflect loudish listening) with 120pf capacitor and measure manually across the range as I don’t have the gear for it. Some other things worth measuring are a 20hz sinewave and how much voltage swing you can get before the transformer core saturates also

The sharp rise in the FR might be worth looking into and tame it a little to keep the amplifier side happy whats the DC impedance on the primary windings?

That was fast, Aokman! All measurements made with 100 pF across the output, that feeding into a very high impedance U pad to take it down 56 dB to line level for my measurement ADC running at 192 kHz. I use 5.1k ohm buildout (safety) resistors inside the amp. I just use 100 V RMS as the nominal (100 dB SPL) because it's a convenient number. I also show the headroom in the rest of my posts. Given that it's a DAC (at 192 kHz sampling) and a sampling scope who knows what ringing lurks in the hearts of men . I do show the 20 Hz performance in this series. I'm really not sure what the cure of the HF rise is, but I've decided to live with +1 dB at 20 kHz, damn the torpedoes, full speed ahead. I don't remember the DC resistance, but I did measure the reflected impedance and on the 8 ohm tap of the toroidal transformer it goes down to a miserable 2 ohms at very low and very high frequencies. So I'm using a 2 ohm 100 watt buildout resistor in series with the power amp to protect the amp, which is rated down to just 4 ohms. Did I answer all your questions?

 

[bobkatz]

That was fast, Aokman! All measurements made with 100 pF across the output, that feeding into a very high impedance U pad to take it down 56 dB to line level for my measurement ADC running at 192 kHz. I use 5.1k ohm buildout (safety) resistors inside the amp. I just use 100 V RMS as the nominal (100 dB SPL) because it's a convenient number. I also show the headroom in the rest of my posts. Given that it's a DAC (at 192 kHz sampling) and a sampling scope who knows what ringing lurks in the hearts of men . I do show the 20 Hz performance in this series. I'm really not sure what the cure of the HF rise is, but I've decided to live with +1 dB at 20 kHz, damn the torpedoes, full speed ahead. I don't remember the DC resistance, but I did measure the reflected impedance and on the 8 ohm tap of the toroidal transformer it goes down to a miserable 2 ohms at very low and very high frequencies. So I'm using a 2 ohm 100 watt buildout resistor in series with the power amp to protect the amp, which is rated down to just 4 ohms. Did I answer all your questions?

Oh, Noise. Noise flat 20-20 kHz is 19.81 dB SPL, or 80 dB below 100 Volts. Crosstalk is not fantastic as you will see, but not terrible. Then 19-20 kHz IM distortion which at nominally 100 dB SPl measures 0.113/0.097%. In the IM plot you can see the garbage from the Class D amp. The IM products are not beautiful either, but anyway, the amp still sounds fantastic! As I said, it is clearly a high headroom rival to my Mjolnir. Beats the pants off of any commercial transformer amp I have heard, especially in clean output level. I like its sound, not too "pretty" not "muddy" and definitely punch, less objectively "clean" than the Mjolnir, but I definitely would not kick this amp out of bed.

 

[bobkatz]

Oh, Noise. Noise flat 20-20 kHz is 19.81 dB SPL, or 80 dB below 100 Volts. Crosstalk is not fantastic as you will see, but not terrible. Then 19-20 kHz IM distortion which at nominally 100 dB SPl measures 0.113/0.097%. In the IM plot you can see the garbage from the Class D amp. The IM products are not beautiful either, but anyway, the amp still sounds fantastic! As I said, it is clearly a high headroom rival to my Mjolnir. Beats the pants off of any commercial transformer amp I have heard, especially in clean output level. I like its sound, not too "pretty" not "muddy" and definitely punch, less objectively "clean" than the Mjolnir, but I definitely would not kick this amp out of bed.

Notes on construction: I fiddled with gas discharge tubes for overvoltage protection and quickly rejected them. When they start to glow, well below their nominal clamping point, they clamp down on the 5.1k resistors, which being 2 W, start to heat up... oops. None of the resistors were damaged but I got rid of the GDTs quickly. Now I'm relying on the amp's nominal clipping point being at a voltage that would produce 1100 V RMS on the secondary of the power transformer. I wager that's safe for my Audeze CRBNs and I've also played my Stax 007 Mk 2's loudly with the amp and there is no apparent danger that I can see, hopefully even from some too loud accidents. Ask me again in a year

 

[aokman]

That was fast, Aokman! All measurements made with 100 pF across the output, that feeding into a very high impedance U pad to take it down 56 dB to line level for my measurement ADC running at 192 kHz. I use 5.1k ohm buildout (safety) resistors inside the amp. I just use 100 V RMS as the nominal (100 dB SPL) because it's a convenient number. I also show the headroom in the rest of my posts. Given that it's a DAC (at 192 kHz sampling) and a sampling scope who knows what ringing lurks in the hearts of men . I do show the 20 Hz performance in this series. I'm really not sure what the cure of the HF rise is, but I've decided to live with +1 dB at 20 kHz, damn the torpedoes, full speed ahead. I don't remember the DC resistance, but I did measure the reflected impedance and on the 8 ohm tap of the toroidal transformer it goes down to a miserable 2 ohms at very low and very high frequencies. So I'm using a 2 ohm 100 watt buildout resistor in series with the power amp to protect the amp, which is rated down to just 4 ohms. Did I answer all your questions?

From what I experienced, playing with various transformers, its generally around the winding resistance and amplifier interactions. You can shift around and manipulate the frequency response by using a series resistor on the primary side and playing with some parallel resistors on the output taps if you need to go even further. It’s basically creating RC / RL networks. You might be able to level out the LF rolloff and HF rise. I prefer not to use resistors at all personally but depends on the transformer.

You will be shocked what even a 1ohm resistor series primary winding resistor will do and completely flip things around.

Load a 20hz sinewave and run up the volume while watching the scope and see what sort of voltage swing you can get also before the sine wave falls apart (minding what you’re doing haha) that seems to be where transformers show their capabilities and worst case scenario for how much voltage drive you have

Transformer input protection is interesting and something STAX seemed to struggle with also (those garbage thermistors). I have been playing with some ideas of zener clamps / voltage monitoring / MOSFET switches and its possible but my first iterations of all this were to maintain the purest signal path possible with minimal component count and just being mindful of the power available and keeping gains limiters in place in my signal chain.

I have implemented zener crowbars on the bias side though to end the show if any overload occurs on that side. Which should be impossible as is with the voltage regulator safety features.

 

[jody2k]

Notes on construction: I fiddled with gas discharge tubes for overvoltage protection and quickly rejected them. When they start to glow, well below their nominal clamping point, they clamp down on the 5.1k resistors, which being 2 W, start to heat up... oops. None of the resistors were damaged but I got rid of the GDTs quickly. Now I'm relying on the amp's nominal clipping point being at a voltage that would produce 1100 V RMS on the secondary of the power transformer. I wager that's safe for my Audeze CRBNs and I've also played my Stax 007 Mk 2's loudly with the amp and there is no apparent danger that I can see, hopefully even from some too loud accidents. Ask me again in a year

Interesting do you notice much improvement pro bias (580 volt) vs boosted 1100v? Just curious

 

[NCUS]

All this activity on transformers is interesting. I have a Krell Evo 300wpc amp, and I have thought about trying a transformer box with it, but have concerns about the Krell blowing up my headphones.

 

[leManu]

Adhesive tape

There is no room to put any tape, the leather pads are in the way. Double sided maybe? Would new pad solve the issue?

 

[aokman]

All this activity on transformers is interesting. I have a Krell Evo 300wpc amp, and I have thought about trying a transformer box with it, but have concerns about the Krell blowing up my headphones.

As long as you’re sensible with it and have a good preamp, don’t hot plug anything, you should be fine. I have used a Sony TA-N9000ES power amp and that was 275W/pc and my current daily driver Sansui 907MR is 200W/pc but it has volume control lol.

 

[gammi]

All this activity on transformers is interesting. I have a Krell Evo 300wpc amp, and I have thought about trying a transformer box with it, but have concerns about the Krell blowing up my headphones.

I use 300wpc monoblocks with my transformers no problem. You can digitally attenuate it first if you want but then the issue there is if you accidentally turn that off, boom. So I just run it normal and make sure I always turn the volume down on the pre-amp first.