Distortion in Janus Preamp with volume past 7

[Oregone]

I have a 73 Rhodes Suitcase with the Janus slider preamp. I have owned this Rhodes for several years (without the amp) and have always used the output of accessory jack 1 with no issues. I recently build a power supply for the preamp, with parts from avion studios. Everything fired up and it appears to be functioning as it should.

That is until I turn it up past 7. Between 7 and 10, there is an increasing amount of distortion. It sounds great when you really play with some spunk, but I am not convinced that this is appropriate or that it will last in this condition. My guess is that it will give up in the middle of a show.

Anyhow, where do I start? Its not my keyboard amp, its not the harp, its not the 5 pin cable, its not the instrument cable, its not the RCA able between the harp and preamp. It is in the preamp somewhere. As I am typing this, i realized that I have not confirmed if it is both L and R outputs. Ill try that at lunch. Assuming that it is not, where do I start? I'm handy with a soldering iron and a multi-meter. I'm in sort of a rural part of Oregon, so taking it to someone is not an option. Id rather diagnose/repair myself anyhow. Thoughts?

[Oregone]

Update: I tried both L and R channels, and the issue is present in both sides.

[vanceinatlance]

Since you are using the preamp with an external amplifier and not the Janus amp, you might be feeding a level to hot to your amp which is causing distortion? I am not used to feeding the preamp into other amps so someone who has been doing that would have better suggestions. Maybe describe the amp you are using?
Vance

[Jenzz]

Hi .-)

If your pickups are adjusted for a hot output (close to the PUs), distortion beginning at around 7 on the scale could be normal, due to the internal gain structure of the preamp. The Input stage has a gain of 22, the output stages add an additional gain of 2,2. Assuming an average harp output of 0,5Vpp on hard strikes, the output voltage of the pramp will be at around 25Vpp !

If you additionally increase the treble or bass, this voltage will increase rapidly, so distortion will occur.

Does the Avion supply +/-15V or maybe +/-12V? Did you check that?  If the preamps runs on +/-12V, ist pretty clear that distortion will appear earlier because of the lower headroom....

Jenzz

[sean]

The Janus preamp isn't too complex, and the schematic is available, so you should be able to track this down.

Your preamp should match one of these schematics:
Schematic for Janus preamp with sliders: http://www.fenderrhodes.com/org/ch11/fig11-1.jpg
Schematic for Janus preamp with sliders, version 2: http://www.fenderrhodes.com/pdf/late-mark1-suitcase-janus1.pdf
Schematic for Janus preamp with five knobs: http://www.fenderrhodes.com/pdf/late-mark2-suitcase.pdf
(The part number on the bottom of the board and the assembly number silk screened on the top of the board whould be a digit or two off from the drawing number on the schematic.)

I will roll the dice and assume that you have version 2 or the Preamp with sliders.

The things I would suspect is that the big electrolytic capacitors are faulty, the 100Ω resistors could be shot, or one of the OpAmps has been damaged.  The good thing is that your preamp is working, so you can monkey around with it while pumping audio through it.

First thing I would test is that the supply voltages at pins 4 and 8 of the op amp chips all read something around 12V or more.  The 100Ω resistors (R20, and R22) might drop as much as 2 or 2.5V before the juice hits the op amp chips (U1 and U3).  Hopefully, the voltage at Pin 4 will be about -13V and at pin 8 it will be +13 volts.  If it is, that's great.  If one voltage is remarkably low, you need to find out why.  Either a 100Ω resistor is actually much higher, or the 100uF caps (C13 or C15) are so shot that they are allowing current to flow.

The old 4558 op amp chips could not provide rail-to-rail output, but they could get pretty close.  If one of the supply legs is low, you would have reduced headroom.

Since you are going to have to solder something to fix the problem, you might as well plan on replacing the electrolytic capacitors.  You can order the repair kit from Vintage Vibe, or order individual components from Mouser or DigiKey.  There are a few 100uF electrolytics to get, and a bunch of 5uF non-polarized electrolytics.  If I ever re-cap my Preamp, I might replace the 5uF (and the 1uF) caps with multi-layer ceramic capacitors. 

I think that if you replace the 100uF caps, the 5uF NP caps, that 1uF NP cap (C16), and check the voltages you will be set.  But if the supply voltages are okay, and the caps are healthy, and you still have distortion, you could have a faulty op amp chip. 

Since your preamp mostly works, you can probe around to find out where the distortion starts.  Rig up a wire to go to the input of your amp, turn the volume to something low, and touch the wire to pin 1 of the first op amp chip.  The first op amp, U1, is at the far left of the board (if the board is upside down, and the knobs point away from your belly).  Then test the output from pin 7 of that same chip.  (It would be easier to probe with an oscilloscope, so it is worthwhile to ask around.  If you have a friend that works at a large auto repair place, one of the tech's might have a modern oscilloscope.  The other musicians in the area must know a good electronics tech.)

If the output at pin 1 is distorted, either the op amp needs to be replaced, or the feedback resistor is higher than 220KΩ, or the input resistor is less than 10KΩ.  (In your case, this is not where the problem is, because the problem is AFTER the volume pot.)  If the output at pin 1 is good, but the output at pin 7 is distorted, either the chip is faulty, the volume pot is faulty, or something is way out of whack with one of the potentiometers (or other components) in the EQ section.  I think the most likely suspects here would be the chip itself, or the volume pot.

If the volume pot had a crack in the element (right around 7), and the ground leg became open as you crank the volume, you could overload the EQ section.  (Yeah, but then why would it work fine when turned down low?  I think a failure that behaves as described is unlikely.)

Since you say that the problem affects both channels, I suspect we have found the problem by now.  Buy maybe not, let's see how far you get with this.

If I were to replace the op amps in the audio path, I would not order MC4558 nor RC4558.  I would use NE5532AP, because of the faster slew rate, higher bandwidth, and the expected better accuracy. 

At the end of all this, I must say that I have never run my Janus preamp with the volume as high as ten, but I do run it at seven and eight most of the time without a problem. 

I am curious to find out how your testing goes.  Keep us posted.

Sean

[Oregone]

Hi .-)

If your pickups are adjusted for a hot output (close to the PUs), distortion beginning at around 7 on the scale could be normal, due to the internal gain structure of the preamp. The Input stage has a gain of 22, the output stages add an additional gain of 2,2. Assuming an average harp output of 0,5Vpp on hard strikes, the output voltage of the pramp will be at around 25Vpp !

If you additionally increase the treble or bass, this voltage will increase rapidly, so distortion will occur.

Does the Avion supply +/-15V or maybe +/-12V? Did you check that?  If the preamps runs on +/-12V, ist pretty clear that distortion will appear earlier because of the lower headroom....

Jenzz

The avion PS is +/- 15V. I tested this at the output of the PS, as well as at the board on the preamp when plugged in and powered on. I am in fact getting the +/-15V.

I believe you may be onto something there with the pickups being too hot in their adjustment. When i voiced and set the pickups on this Rhodes, I was taking the signal directly from the harp. I am going to try adjusting a couple of pickups back and see if that alleviates the issue for those notes. This was something that I never considered.

[Oregone]

Since you are using the preamp with an external amplifier and not the Janus amp, you might be feeding a level to hot to your amp which is causing distortion? I am not used to feeding the preamp into other amps so someone who has been doing that would have better suggestions. Maybe describe the amp you are using?
Vance

The amp that I am using to test this is a little Fender Sidekick Keyboard. It has channel level, as well as a main volume. Regardless of the setting of either volume pot (even down to barely audible) i have the same distortion. I dont think that I am over-driving the keyboard amp.

[Oregone]

I will roll the dice and assume that you have version 2 or the Preamp with sliders.



At the end of all this, I must say that I have never run my Janus preamp with the volume as high as ten, but I do run it at seven and eight most of the time without a problem. 

I am curious to find out how your testing goes.  Keep us posted.

Sean

Thanks for all of that info! Just what I was looking for....

I do have the version with sliders.

When I was initially testing my power supply, it was later at night and the kiddo was asleep. I didn't turn it up past 3 or 4. But the next day i tried turning it all the way up, and that's when I discovered the issue. I doubt I would play it on 10, but in the heat of the moment at a gig, i don't want to be concerned with making sure to not turn the volume up too much.

I'm going to try adjusting a few pickups out this evening, and seeing if I just had them set too hot. If not, I will start with your diagnostics. I will have an update by morning. Thanks!

[Jenzz]

Hi .-)

Looking at several Sidekick schematics, it's possible that it's the input stage of the amp that distorts. The volume pot works as a gain adjustment for the Input stage, so the pot doesn't affect the signal itself. If the input level gets to high, distortion will occur no matter how low the pot is turned.

Jenzz

[pnoboy]

Hi .-)

If your pickups are adjusted for a hot output (close to the PUs), distortion beginning at around 7 on the scale could be normal, due to the internal gain structure of the preamp. The Input stage has a gain of 22, the output stages add an additional gain of 2,2. Assuming an average harp output of 0,5Vpp on hard strikes, the output voltage of the pramp will be at around 25Vpp !

If you additionally increase the treble or bass, this voltage will increase rapidly, so distortion will occur.

Does the Avion supply +/-15V or maybe +/-12V? Did you check that?  If the preamps runs on +/-12V, ist pretty clear that distortion will appear earlier because of the lower headroom....

Jenzz

The avion PS is +/- 15V. I tested this at the output of the PS, as well as at the board on the preamp when plugged in and powered on. I am in fact getting the +/-15V.

I believe you may be onto something there with the pickups being too hot in their adjustment. When i voiced and set the pickups on this Rhodes, I was taking the signal directly from the harp. I am going to try adjusting a couple of pickups back and see if that alleviates the issue for those notes. This was something that I never considered.

I wouldn't pull the pickups back--that's letting the tail wag the dog.  The early Janus preamps had a lousy design for the 1st stage--as a previous poster mentioned, the gain of that stage is awfully high, and could, under some circumstances possibly saturate.  Also, its input impedance is very low (10k).  You could potentially fix both problems by replacing the first 10k resistor with a 20k resistor.  That will cut the gain in half and also let more high frequencies of the harp through.

[Oregone]

I had limited time to work on this over the weekend, but I did have some.

I started by pulling one octave of pickups back. Naturally this resolved the issue.
They are all set pretty close to the end of the tine, and I moved them near the middle of their adjustment. I could turn the preamp all the way up, with no distortion.
My first thought was to dig in and re-adjust the entire piano. Common sense took over, and for the sake of time, i just decided to keep the volume rolled back rather than re-adjust the entire instrument.

I still intend on going thruough all of the troubleshooting steps provided by Sean (thanks!) when time allows.
I may end up replacing the 10k resistor with a 20k as suggested, if no other issues arise.

Thanks for all of the words of wisdom, I will continue to update this thread as I address the issue further.

[sean]

The first preamp stage is not to blame here.   If the first stage was clipping, then turning the volume below 7 would not clear up the distortion.

Since pulling the pickups back provided some satisfaction, maybe the offensive noise is not the fault of the preamp.  The bass notes can provide some fat flabby buzz if the pickups are pulled in close, and moving the volume knob doesn't clear it up (hitting the keys softer does clear it up).  I believe this sound is due to saturation of the magnetic core in the pickup, but it could be the fault of having so many inductors in series the wild swing of the long bass tines.

With the pickups in close to the end of the tine, and a good strong hit on A13, the output looks like:



This oscilloscope screenshot pretty clearly shows that the output is not nearly as sinusoidal as ideal.  There is a pretty steep spike on this note, and it sounds like a fat flabby buzz that eventually goes away as the note fades out.  When the pickup is moved back a bit, the steepness of the attack smooths out.


There is evidence of an inductive kickback that distorts the ideally sinusoidal output from the pickup.  The resulting waveform has an additional lobe or lump in each cycle, or strongly asymmetrical north and south lobes. 

A strong hit on E56 shows one lobe is much smaller than the other:


I like the distorted sound I get in the bass notes when I hit them hard.

Sean

[pnoboy]

The first preamp stage is not to blame here.   If the first stage was clipping, then turning the volume below 7 would not clear up the distortion.

Since pulling the pickups back provided some satisfaction, maybe the offensive noise is not the fault of the preamp.  The bass notes can provide some fat flabby buzz if the pickups are pulled in close, and moving the volume knob doesn't clear it up (hitting the keys softer does clear it up).  I believe this sound is due to saturation of the magnetic core in the pickup, but it could be the fault of having so many inductors in series.

The core saturation causes a pronounced kickback that distorts the ideally sinusoidal output from the pickup.  The resulting waveform has an additional lobe or lump in each cycle, or strongly asymmetrical north and south lobes. 

With the pickups in close to the end of the tine, and a good strong hit on A13, the output looks like:



This oscilloscope screenshot pretty clearly shows that the output is not nearly as sinusoidal as ideal.  There is a pretty steep spike on this note, and it sounds like a fat flabby buzz that eventually goes away as the note fades out.  When the pickup is moved back a bit, the steepness of the attack smooths out.

A strong hit on E56 shows one lobe is much smaller than the other:


I like the distorted sound I get in the bass notes when I hit them hard.

Sean

Your comment about the 1st stage of the preamp is right on--I missed the information that turning down the volume cured the issue.  However, your comment about the pickups producing a pure sinewave is not correct.  The ends of bass tines swing quite a distance when struck hard.  As such, at the extremes of their travel, they are quite far from the pickups, and the pickups' sensitivity to their motion is very low.  As they travel to their mid position, which is close to the pickups, the pickups and tines couple much more strongly, and thus we have the fact although the swing of the tines is sinusoidal, the output of the pickups is not.   Also, I don't see any mechanism for the pickups to become saturated.   Consider the simple case of a single pickup connected to an infinite impedance amp.  In that case, there is no current in the pickup, so the pickup produces no power.  Now consider the real case--there is very little change in the response of the bass pickups between their being directly connected to a 1 meg input impedance guitar amp and being connected to the Janus amp shown or to the stage piano's volume control, both of which have an input impedance of 10 kohms.  At one point, I electrically isolated one pickup from the rest and connected its output only to my amp--there was no particular change in tone.

[sean]

Duh.  The wild swing of the long bass tines is obviously a much better explanation for the crazy waveform.

And I guess the humped asymmetric output at E56 is inductive, but not core saturation.