Wurlitzer 200A: Vibrato hum and distortion

[gacki]

(A cursory search didn't turn up anything, so if this was discussed already please point me to the thread in question. Thanks.)

I'm more of a Rhodes person but was asked by a couple of fellow musicians to work on their Wurlitzer pianos. Recently I already posted about my experiences with a 200; now it's a 200A.
I did a 230V conversion that went pretty flawlessly although I wish the the ep-service.nl transformator had longer primary leads.

The amp is generally sounding fine with the appropriate "bark" and with practically no hum (although the hum shield isn't installed yet). This radically changes when turning up vibrato.

On the 200 it appeared to me as if the vibrato circuit was attenuating the sound; on the 200A it seems as if the sound is boosted instead. Is that correct? It quickly goes into overload and also exhibits lots of hum in those louder portions of the modulation cycle.

Is this normal? Is there something that can/should be done except turning the volume down before turning vibrato up?

[Jenzz]

Hi .-)

There are two trimpots on the 200A PCB...

One locatetd on the left front edge (max. volume adjustment), this one can be left as-is.

The second one (red dial mostly) on the left side in the back, right before the heat sink.  This is to adjust the average volume change / gain when tremolo is dialed in.

Jenzz

[gacki]

Thanks, that helps a bit but not much. Is that behavior perhaps normal?

Even with max volume and no tremolo I don't get any noticeable distortion. But with tremolo even with both trimpots at min I get very nasty distortion that makes me fear for the speakers (it's not as bad as with the original settings but still...)

I'm thinking out loud here: With no tremolo everything is comparatively fine. The output of the preamp is around 1.8V peak to peak for heavy handed playing; the signal looks clean on the oscilloscope. With tremolo at max and the tremolo trimpot at min the preamp signal is around 4V peak to peak for similar playing as before; the signal still looks fine and non distorted on the oscilloscope.

This seems to be the "late model" version with the Aux volume control after the Aux circuit.

When turning up the Tremolo control it looks as follows:

The Aux out (pin 11) clips slightly; this is also obvious in the scope. The Aux in (pin 9) does NOT clip.

The headphone out also clips (so do the speakers, naturally). This looks like hard clipping at the power rails; and for obvious reasons this sounds like a recipe for disaster for the speakers.

So my interpretation of all this would be that both the Aux amp and the power amp operate near their maximum capacity without tremolo. Dialing tremolo in results in slightly more than 6 dB signal gain; and both amps can't handle this.
This is insofar irritating that I can raise the "max volume" trimpot to the maximum when not using tremolo and this only results in a slightly "grittier" sound - but when using tremolo things get really nasty even when the "max volume" trimpot is at minimum position.

I also find the tremolo comparatively weak - but on the other hand I'm more used to the Rhodes Suitcase one. It certainly is very different to the one in the Wurlitzer 200, too.

Is all this really normal and expected behavior?

[Jenzz]

Hi .-)

This is not normal.... Did you check the 15V comimg from the regulator (TP 19 on the PCB)? There should be from 13,5 to roundabout 15,5 volts. If this voltage is low due to a bad regulator chip, signal headroom is decreased and distortion occurs...

Jenzz

[gacki]

It's 14.5V which I consider close enough.

The distortion in the power amp seems to be rail distortion; meaning the signal is clipped at +/-22V (roughly). This would probably sound acceptable with a tube amp but is pretty bad with a transistor amp.
I'm not fully sure where the distortion in the Aux circuit comes from.

(This particular Wurlitzer had blown at least one of its output transistor fuses a couple years back; not sure if there was a particular reason for this.)

Voltages on TR1 and TR2 look ok; and the output of the preamp looks fine to me.

The Wurlitzer has a very prominent marking for the volume knob at about the 2 o'clock position which suggests to me that the owner does not intend to play it any louder. That's roughly the position where the signal stays below the clipping when using vibrato.

Thanks, Malte

[Jenzz]

Hi .-)

14,5V is ok !

What about the value of the Tremolo Pot (50k) ? Is it wired correctly? What happens when you diconnect the wiper wire from TP5?

If TP5 is disconnected, the Tremolo pot should do nothing....

Jenzz

[gacki]

Wiring is correct; values are also within spec. With TP5 disconnected the Tremolo pot does indeed nothing.

This of course points a bit at the vactrol, doesn't it? I've seen vactrols fail before...

Thanks, Malte

[Jenzz]

Hi .-)

Yes, ols vactrols are prone to failiure when getting older... In most cases, the LDR element will drift out of spec.

The original vactrol seems to be not available anymore... I've used the VTL5C3 with good results.


Jenzz

[gacki]

Hi,

and thanks again for your input.

After talking to the owner I'm not fully sure if it's worth tackling this problem because it was something he was used to. But I might give it a try if the owner decides to go "all in". Which leads me to the next questions/problems:

It's clear to me that the hum is mainly originating from the preamp; when desoldering the preamp cable the power amp is producing very little hum (it's faintly audible but in my opinion non critical). Is there something I should do to the preamp? So far I have refrained from recapping anything since the caps look and measure ok.

The other problem is significantly trickier for me - and I didn't notice it until today: While working on the Wurlitzer it naturally was open and the lid was standing next to it. Imagine my surprise when I put the lid on (so far everything ok) and then closed it and the hum became slightly worse...
It's definitely the 50 Hz fundamental that is getting louder - not drastically but definitely noticeable. Could it be that the mains are coupling directly into the speakers? Or is it perhaps the transformer coupling into the right speaker and though it also into the left one?
(This 200A has the 230V transformer from ep-service.nl installed which seems to get about 1 centimeter closer to the speaker when mounting it without modifying anything. It certainly would be nice if there were mounting instructions for it and if the primary leads were slightly longer...)

[Jenzz]

Hi .-)

In most cases, shielding is an issue... There are internal ground loops... The conductive paint inside the case bottom can loose connection to ground... Allways use shielded power cables inside the piano.

When the hood is open, the speakers see no air volumina to work on. If you close the hood, the air volimina inside the piano acts as in a closed speaker box. This will increase volume and bass, so the hum gets louder when the hood ist closed.

Jenzz 

[gacki]

Thank you for your continuing input and ideas. It definitely helps me understand the whole construction and behavior.

Let's reorder this a bit:

When the hood is open, the speakers see no air volumina to work on. If you close the hood, the air volimina inside the piano acts as in a closed speaker box. This will increase volume and bass, so the hum gets louder when the hood ist closed.

So it's essentially the "geschlossene Box" principle. Makes sense.

In most cases, shielding is an issue... There are internal ground loops... The conductive paint inside the case bottom can loose connection to ground... Allways use shielded power cables inside the piano.

The ground loops are definitely there (and getting rid of them changes the hum somewhat). I'm trying to find out the reasoning behind the way the ground wires are connected and where to break the loop(s). Here's my understanding:
The main problem is of course the ground for the tone bar: It's grounded to the amplifier chassis through the preamp (the screw on the bracket connecting both halves of the tone bar) and the grounding wire from the amplifier to the "star ground". But it's also grounded to the amplifier chassis directly through the right side ground wire and (somewhat?) through the left one. Removing the right side ground wire is cleaning up the sound (especially removing the fundamental of the hum), so this one probably should be disconnected. Removing the left ground wire has no audible effect.
I'm a bit puzzled how the two "brackets" onto which the tone bar and action are mounted are supposed to be grounded. As far as I can see the left bracket is automatically grounded to the amplifier chassis because the latter is directly screwed onto it (with the larger screw). So what is the left wire (between the bracket and the tone bar) even supposed to accomplish? On the right side things are a bit more complicated: At least on the model I'm working on the bracket is practically "accidentally" grounded to the amplifier chassis because it simply rubs against it. Still it probably would be a good idea to have a permanent ground wire run to the back of the bracket (as it is stock)?

There's also the question where to run the ground wire from the mains inlet (to the star ground point right of the amplifier board). On this particular model it was simply dangling across the amplifier board. Right now I've routed it through the heat sink but I could imagine it would be best located next to the transformer wires and under the amp board?

Lastly, the conductive paint. I can measure around 100 Ohms at the places where the ink is present but I don't get any connection to the main ground. How is that connection supposed to happen - just by screwing the amp chassis on some wooden (and hopefully paint-covered) parts? Sounds a bit problematic from an engineering perspective.

(I've also re-routed the output cables away from the power cables which poses another problem: What is good practice for routing power and audio on the left bracket? For obvious reasons they are somewhat close there. Right now I've attached the power cable to the outer side and the audio cables to the inner side.)
 
All in all the sound definitely has cleaned up already so I'm probably trying to address the the distortion problem (ordering a vactrol now); and I also might rebuild the preamp.

[gacki]

So here's the current situation:

Changed the vactrol to a VTL5C3. This definitely makes a clear difference: The "vibrato" is no longer as strong as before and the distortion is much lower as well. I adjusted the vibrato trimmer a bit so that the intensity is somewhat increased again. In general this seems to be a good compromise between vibrato depth and distortion now. The rail distortion (that was unavoidable with the old vactrol) is no longer there.

Hum and hiss: I've tried to measure the result of various modifications with a spectrum analyzer plugin. Whatever I do there are peaks of the line frequency and its harmonics (50 Hz, 100 Hz, 150 Hz and so on). Interestingly the 50 Hz component is almost unaffected; it's mostly the harmonics.
Signal is always taken out of the Aux out at full volume.

Without any modifications the 50 Hz peak is at around -73 dB, 100 Hz is around -87 dB, 150 Hz is at -79 db, 200 Hz is at -81 dB.

Disconnecting the right hand ground connection to the shield lowers 150 Hz to around -86 dB and 200 Hz to around -93 dB. 50 Hz and 100 Hz stay about the same. This was the most audible change for me.

Changes in the preamp: R1 and R9 to metal film; TR 1 and TR2 to 2N5088, swapped electrolytic caps: No clear change. The hiss seems to have decreased slightly (around 2 to 3 dB).

Recapping C28 and C29 in the power supply: No change; but perhaps a slight decrease in hum from the speakers (which would be normal since those two caps filter the +/-22.5V which is only used in the power amp).

When the Wurlitzer is turned off I'm still getting a peak for the line frequency (but not the harmonics): the peak for 50 Hz is around -105 dB. When I remove the line cord this peak (naturally) disappears. So there must be some interference directly from the wiring. When I move the line wiring between the power input and the molex connector clearly away from the chassis the peak drops to around -109 dB. Unfortunately there isn't much room to do any of this when the lid is closed...

And finally: with the preamp disconnected (desoldered at terminal 3) line interference drops by around 4 to 10 dB; both with the Wurlitzer turned off or on.

(and before anyone asks: the recordings/measurements were done with a Laptop without a power supply and a USB interface so there is no additional potential ground loop)

My question: Is this as good as it gets (aside from installing another amp of course)?

[Jenzz]

Hi .-)

Regarding hiss: The voltage regulator chip (723) is often noisy, because this chip was intended for industrial use, not esp. for audio. In most cases a 100nF film cap across R54 will reduce hiss audibly.

How is your audio measurement setup? It seems there is a ground loop between the piano and your setup, because you have some values even when the piano is off... This can only happen if a ground loop is present.

Jenzz

[gacki]

There is no ground loop between the Wurlitzer and the Laptop that I'm aware of - the Laptop runs on batteries (for the measurements) and the interface is USB powered from the Laptop.

I'll give the hiss fix a try once I got the amp running again - the pre-driver transistor just blew... On the other hand the hiss wasn't too bad; I'm more concerned with the hum.