DCAudioDIY.com

I've been working on my Musician' Amplifiers with Heyboer's Peerless S-265-Q copies, and the end results are extremely good. I finally bought a portable CRT scope on eBay, it was $70 from some lab in Minnesota, in perfect condition. Also got a little Koolertron function generator, as well as a Picoscope 2004A for more sophisticated measurements. I've got square waves down, that was easy. . Hoping to do some XY readings to look at phase shifts and so forth, but for now the amps are in good shape. Pretty much what you'd hope for from an "original" triode Williamson: Coherent, good bandwidth, nice full presentation of instruments and voices, wonderful to listen to--no listener fatigue whatsoever. Frequency plot shows it dead flat out to about 70kHz then a gradual rolloff with nothing sticking out after that. Stability is *pretty* good--HF's are stable with .1uF across an 8 ohm load, and LF looks solid, but I'm hoping to do better by tuning the shelf network on the first stage. I also need to get them off the breadboard, because that isn't helping any, with long leads and hookup wire everywhere.

To give you an idea, first pic is 10kHz with no tuning at all. Yikes! Second pic is 10kHz after adding standard Williamson shelf network across the first plate resistor, and adding phase lead cap across the feedback resistor. Third pic is 100Hz response. I want to improve stability without ruining the sound of the amp, so I'm going to tinker some more with the shelf network. I've tried some extreme methods but while stability is very good, the amp sounds unnatural--the frequency balance is off and the amp calls attention to itself, not what you want.

I'll post a modified Musican's Amplifier schematic shortly, and look forward to any suggestions. But I'm very pleased with the results.

Well, it's been a bit of adventure lately. I ordered a new pair of Heyboer/Peerless trannies, this time with 50% primary taps to more closely duplicate the originals and allow for a "Gilded Lily" or "ultralinear" version of the Musician's Amplifier. They arrived in due time, but there was a problem:



Yikes! Here's the open-loop response:



Obviously they really screwed something up. Back they went, and after a lot of discussion, they wound a new pair:



That's more like it!

(Cont.). Here's a FR plot with a standard Williamson shelf network across the input stage (4.7K + 200pF) and 150pF across the feedback resistor:



Meanwhile, as a kindly fate would have, I found two original Peerless S-265-Q's on eBay within three days of each other, for a not-unreasonable price, so I can now directly compare the copies with the originals. Here's the same set-up with an original 265-Q:





So you can see the Heyboers are not very far off at all. Sonically, the originals have a very silky top end, as you might expect, while the Heyboer copies are less refined but still very clean and musical. On the plus side, the Heyboers sound equally good in triode (11wpc) or "ultralinear" (22wpc) mode, while the Peerless give up a little bit of realism in ultralinear mode. It's "pretty" but it doesn't have that triode solidity and immediacy.

Would I choose the originals over the copies? Yes, I would, but I'd be perfectly happy listening to the copies as well. And the originals very rarely come along. It was sheer luck to find the pair that I did. But if you have patience and persistence, they do turn up now and then, and they're worth getting if you can. Considering that the Sowter "Williamson" OPTs cost $900 plus shipping to the US, and the Hashimotos about as much if not more, it's not unreasonable to pay $300-400 for one. Meanwhile the Heyboers are about $450 a pair including shipping, and they make a swell Williamson amplifier.

Here's a schematic for the Heyboer Musician's Amplifier. This is quite an amp, if I do say so myself. Direct comparison with the original Peerless shows that the Peerless are more silky on top, and more liquid overall, but the Heyboers are punchy and crisp. I could live with either one. The "ultralinear" version gives up very little to the triode version, and yields a very robust-sounding 22wpc. I cannot recommend this new/vintage circuit highly enough.

If you want a bit more power, use a good 6L6GC or, even better the new Tung-Sol 7581a, and the 5V4 rectifier with the 400-0-400 power tranny. This gets you a bit more punch and does not stress either output tube.

Please explain the resistor/cap networks to ground

Those are just power supply decoupling caps. They're stacked electrolytics with equalizing resistors across each caps. I drew them in that way to emphasize that you must have plenty of voltage headroom in the power supply capacitors.

The original Musician's Amplifier used 600vdc electrolytics, which you cannot find anymore.

You could use DC Link film caps with 600 or higher voltage limits.

https://www.mouser.com/ProductDetail/WIMA/DCP4I052507GD2KYSD?qs=sGAEpiMZZMv1cc3ydrPrF77Ed2j2AojdDai0LtMjM6g%3D

The two 47 uf caps in series could be replaced by a single 25 uf and you wouldn't need the resistors.

ray

There are some currently manufactured 600v electrolytics that are available at places like Digikey and Mouser. Search for Nichicon LGN series capacitors and you'll find values ranging from 56uf to 220uf all at 600v rating. I haven't tried them but Nichicon typically makes high quality caps.

Yes, you could. Even Solens would do the job. I haven't cared for the sound of the plastic caps I've tried but I'm always willing to experiment.

Interesting, I'll take a look. The first cap at the rectifier needs to stay low, 8 or 10uF, to keep the rectifier from freaking out, but after that I haven't read anything that indicates that increasing the decoupling cap values a bit leads to any instability. I did increase them a bit to the values employed by Heathkit in their version of the Musician's Amplifier, the W-2, mostly because it's easier to implement.