!!!ADDENDUM!!!
!!!WARNING!!!
The voltages in the following schematics are potentially lethal! Build at your own risk!
![]() |
| Dynaco floating para-phase inverter |
As I had mentioned in my Western Electric 131A/133A amp article, the very first amp I built from scratch was the above floating para-phase schematic, using transformers I scavenged from a rusty Dynakit SCA35. That was a nice-sounding amp that I've always wanted to revisit despite some trepidation. I was held back by the global negative feedback (1k + 0.001ยตf) from the output transformer's secondary back to the input tube's cathode.
![]() |
| Western Electric 124 |
After many decades of DIY'ing tube preamps and amps, by far the only type of feedback I've encountered, which didn't get in the way of my musical pleasure has been the local feedback illustrated in the above WE124 amp schematic: C3/R13 from plate to cathode and C6/R14 from plate to grid. This local feedback topology can be found in various preamp and amp models produced by Western Electric. Being a floating para-phase inverter circuit using two pentode 6J7 tubes was another incentive to build this instead!
See the HiFi Haven WE124 amp and WECo 133A threads.
May 4, 2026 Addendum
It turned out that all the test data I presented using the Eico ST70 output transformer were done with the amp circuitry wired in reverse phase.๐ฒ
My apologies! ๐
![]() |
| 30 Hz - 2 kHz = ruler flat, +1dB @ 3.5 kHz, +2dB @ 4.5 kHz, +3dB @ 6.5 kHz, peaking at +4dB from 9 kHz to 18 kHz. Then it rolls off 3dB @ 19 kHz to 0dB at 25kHz. |
A few days after this article was uploaded, I installed another transformer (pulled out of a Scott 99D amp) on this amp chassis. That's when I realized mistakes in my wiring. After correcting my errors, I thought the amp sounded even better but when I put the amp on the test bench, the square waves and frequency response turned out to be wild! ๐ฒ
![]() |
| Test data with the Scott 99D output transformer ~ 7600ฮฉ primary Z, C12 = 50pf cap and correct phase ๐ |
With a 0.236V rms input, output power is 6W into 8ฮฉ before clipping. The square waves were taken with a 50 pF cap soldered across both plates of the 6J7s. Frequency response (taken with 2.7 V rms output into 8ฮฉ) was ruler flat from 30 Hz - 1 kHz, then starts rising to a 2dB peak @ 5 kHz, slowly rolling off to 1 dB @ 10 kHz and -2dB @ 15 kHz. This might not be acceptable to Hirsch-Houck Laboratories, but at JE Labs it's about musical satisfaction! ๐
![]() |
| WE124-style PP6V6 amp + Eico ST70 output transformer on a GepettoCoLLC 12" stool ๐ |
As mentioned earlier, I prefer the higher primary Z output transformer from the Scott 99D. It's more focused, detailed yet harmonically rich and airy. The much chunkier Eico ST70 iron might produce more low-frequency energy. But it may also be psychoacoustic due to its more reticent top end.๐
In my WE131A/133A derived amp, the sonic difference between the pentode 6F6 and 6V6 beam tetrode wasn't as pronounced. However, in this amp, the 6F6 sounded rather unexceptional. Perhaps this is just another indication that the WE124 circuit was optimized for beam tetrode output tubes. But I like the ๐ถ sounds I'm hearing from the 6V6s and that's what matters most!
![]() |
| WE124-style PP6V6 amp + Scott 99D output transformer |
The tonal balance of this amp is reminiscent of my Radiotron SE2A3 whereas the WE131A/133A derivative is akin to my nickel plate choke loaded SE2A3 amps. Both circuits allow the harmonic richness of the 6J7 pentode to shine through, so I don't think it'll be an easy choice. Since I've only had this amp running for barely a couple of weeks, these are just my initial impressions. I look forward to going through my LP and CD collections to further familiarize myself with this amp.
It would've been ideal to compare these two WE-inspired amps using the same output transformer. Anybody willing to donate a pair of Dyna Z565s? ๐
You'll know which route I decided to take when I upload the stereo amp build.๐








I recently built a PP all triode amp (GU-50 and 6E5P) using Eico ST-70 outputs. My 10k square wave is much cleaner than shown above with the 6V6. I suspect the pentode output tube, even with some local feedback just can't drive the big Eico transformers. And they are probably a bit low for primary impedance. 6L6's might do better. Enjoy your blog. Thanks.
ReplyDeleteIME, triodes usually show very nice square waves even without feedback. I just replaced the EICO ST70 output transformer with a ~ 7600ฮฉ pri. Z pullout from a Scott 99D. Sounds even nicer to my ears but looks worse on the scope. ๐
DeleteThanks for visiting and the kind words!