Wurli 200A Refurb - Reducing Hum, Noise & Improving Transistor Bias

[JamieOxford]

Hello, I'm refurbing my Wurli and looking for some advice on fine tuning the electronics and noise.

I've just given it an overhaul and everything's technically working. It sounds a lot better than it did but there's still room for improvement. I've replaced all preamp components, electrolytic caps and big red caps with a mix of bipolar, decoupling, film caps, metal film resistors and modern transistors, (all matched to original specs).

R58 was removed & R34 increased to 199Ω to achieve +0.57V at TR9 (originally +0.35V), although I still only have ~1mV across R37. All nearby resistors and test points seem to be in spec (with 25.6 & 25.17V at TR11 due to increased rail voltages). Christian at Tines & Reeds suggests this should be 8-10mV, so should I keep increasing R34 to achieve that or look at modifying R36 maybe?

Overall noise and hiss seems better but is still audible. On full volume, loud notes are ~25V peak to peak, quiet notes are ~10V peak to peak, the hum on the scope is around 40mV peak to peak and noise spikes are around 200mV.

Internal voltages differ from spec a bit, but don't seem totally out of range:
+/-25.8V, +142.6V, +14.12V

One thing I've been trying to get rid of is the small 'pop' on the speakers a few seconds after powering off, but this could be bias related. When this happens the +25 rail drops to +0.5V whereas the -25V rail stays at that voltage. It looks like the - cap is still holding charge at this point whereas the + cap is being discharged but not sure if that's normal? Disconnecting the neon lamp doesn't change the hum or pop either.

There's also some occasional HF noise in the signal path which could be new capacitors and transistors settling in, or could be bias related again. It's prominent in the decay of some notes, especially in the bass region but sometimes sounds like it's quietly always there in the background.

C28/29 were changed to high spec Vishay 056 PSM-SI 2200uF caps. These were the best I could find that would physically fit the board correctly and I'm guessing there wouldn't have been a lot of benefit in increasing the value here?

The vibrato peak to peak voltage is 9.76V whereas the schematic suggests it should be 11.5V. If the 15V rail was correct, I think this would still be a bit lower than it should be (assuming the same current draw).

I'm also considering star grounding it to reduce the hum. If I measure the resistance between different ground points across the piano, the value jumps around a bit so I'm assuming there are some ground loops in there. The mains outlet is earthed and it looks like the 0.1uF cap by the neon lamp was never installed or has been removed (assuming this was the 'death cap'?). The mains wires also have their own metal channel behind the heatsink to keep them away from the amplifier.

Another anomaly is the Aux output has a 680Ω resistor in series with C26 - it looks factory spec and I'm assuming it's to limit current/output voltage? (Just wondering as it's not in the schematic)

Regarding my earlier post, I replaced the speaker fuses with fast blow ones and they're working well. (One fuse blew previously as a transistor leg had broken before the refurb, but otherwise was all working ok.)

The only potential issue I can see from my refurb is I put a bipolar cap at C12, but not sure if this would benefit from being a standard electrolytic instead? Also, I've been working without the hum shield on but installing it makes no difference to the noises mentioned above.

[velo-hobo]

Personally, I don't rely on google ai search results, especially for relatively obscure questions. I've seen it return a good many inaccurate or misleading results.

If I was working on that amp, first I'd ask myself why that fuse is blown. Is something else in the power amp/output section shorted out and pulling too much current?

Looking through a scanned copy of a 200A service manual, the parts list does not indicate fast or slow for those two fuses. However, the line/mains fuse is explicitly indicated as slow-blow in both the parts list and the schematic drawing. So my *assumption* is that if F1 and F2 are supposed to be slow-blow fuses, then they would be labeled as such.

Did you pull the old fuse out to examine it? Not that we can be sure it is original or correct, but I'd at least have a look to see if it was even of the correct amperage rating.

[JamieOxford]

I made some progress today by star grounding the Wurli which has reduced hum quite significantly. You can see in the spectrograms that the 50Hz cycle and harmonics are much quieter though there is still some 50Hz when the volume is turned up (which I expect must be in the preamp section). These tests were done with cheap test hooks for the star ground connections and without the hum shield in place, so hopefully it'll be even quieter once that and the proper earthing cable are installed.

Can anyone comment on whether it would be worth upgrading to 4700uF smoothing caps for the power supply? Nichicon have discontinued a lot of their audio range and the only substitute I can find of the right diameter are these 'Audio Note' ones which are supposed to be good.

I'm also wondering if there's any merit in increasing the value of C30, 31, 34 or swapping any to electrolytics to reduce the 50Hz cycle in the preamp?