I've been thinkin' about how Robert Keeley modifies Boss CE-2s so they can be used with non-Boss power supplies.  From what I can tell, Boss pedals only have 10v rated electrolytic caps or something, and that's why the pedals fail when fed higher voltages...right?  Does that mean that Robert gets around this by replacing all the electrolytics with caps that have a higher voltage rating?

The pic he has on his website looks as if the hi-fi mod is as simple as replacing the 1uF electros with films, putting in a BB opamp, and replacing the power filter caps with 50 or 63v electrolytics.

Wasn't he talking about the ACA to PSA reg. power supply mod? I have two MIJ CE-2's that are ACA. Tomorrow I'll check the electro's to see what the voltage rating is. As long as they're 16v or better, it should be fine. I'm gonna try his "hi-fi" mod, replace some caps and resistors with metal film. Thanks to TI, I've got OPA2064's & OPA2134's to try. Maybe try an NE5532.

http://www.jpl-audio.com/images/acapsa.pdf

Rule of thumb is that caps having anything to do with power supply should be rated at the value closest to double the supply voltage.  Given that there is typically no rating between 16v and 25v, the closest value to 2 * 9v is 16v.  I can't imagine BOSS designers putting in anything less than that when it comes to supply and bias voltages.  Certainly space-savings does not require it now or in 1982 and cost-savings are likely negligible if they exist at all.

When electrolytics are in the signal path, that's another thing.  These can be rated a great deal lower and a 10v rating on a series cap in the signal path that won't see anything more than a volt or so is perfectly reasonable.  I don't think anyone would expect signal levels within a CE-2 circuit to be much more than a volt.  The sole exception would be if the signal is superimposed on a DC bias-voltage; e.g., 500mv riding atop a 7v DC bias.  So, generally speaking, the required voltage rating for an audio signal going into a gain/processing stage can be quite a bit lower than the voltage rating of those caps smoothing out the power supply.  Of course, if the audio input signal is then placed atop a DC voltage, whatever cap is parked at the output of that stage needs to have a high enough voltage rating to accommodate that extra DC.

In the case of the CE-2, and likely any BBD-based modulation device, there is a bit more to adopting a higher supply voltage than simply assuring the caps will remain on spec.  Bear in mind that the trimpot for setting the bias voltage to the BBD doesn't "know" what voltage it is feeding the BBD input.  It is simply set to some *proportion* of the supply voltage, based on an assumption of what the supply voltage is, and with a 9v battery or standard 9vdc wallwart, you have a pretty good idea what that is.  Switch to another supply voltage, though, and although the MN3007 *can* run with it, the bias setting isn't necessarily correct with the new supply voltage and may have to be tweaked.

Similarly, the clock circuit formed around the MN3101 takes its feed from the LFO, whose rate and amplitude, in turn, will depend partly on the bias voltage being fed into the inverting pin (pin 6) of the dual op-amp used for the LFO.  I haven't tinkered with one, but it stands to reason that the LFO will perform a little "differently" with the new supply.

All of these changes resulting from the new supply voltage CAN be adapted to, so it is not impossible.  One just has to DO the adaptation.  

In the meantime, depending on what sort of supply voltage you plan on using, determine if the relevant caps will need changing for safety's sake.  If the intent is to use something closer to 15v, then moving from a 16v rating to 25v is probably a smart thing to do so you can blithely carry on assuming the caps are on spec.

Thanks Mark.  I'm just assuming that he means the CE-2 can be used with higher voltage power supplies to get more headroom.  The way Robert explains it led me to believe it was just a cap substitution.

He describes the mod as:
"Power supply filter modded so that standard power supplies work better. Higher voltage leads to better headroom"

...and this picture is subtitled:
"Note the larger voltage capacitors"

...and I'm also curious as to whether adding a burr brown or RC4885 chip and replacing the electro caps in the signal path with film would make a big enough difference to be worth trying it out on other modulation pedals.

Thanks also Mark, quite a detailed explanation. I assumed Keeley was just talking about the ACA to PSA mod, but upon reading it again, I see he was talking about higher voltage supplies. I think I'll just convert mine to a regulated 9v PSA type supply, and try some metal film substitutions and different op-amps.
BTW, if it was converted to a higher power supply, it seems that the trimpot setting would be only correct with the power supply, not with batteries, correct? Thanks.

The ACA power supplies are dirtier power but they run at a slightly higher voltage than 9v like batteries.  Boss added a little resistor here to lower the voltage when using a power supply.  If you simply short this resistor out, then a regular 9v power supply will work.  

I found this out by using a Visual Sound 1Spot on my Japan CE-2 and the light was lower than when I used a battery with it.  Also, it didn't sound as good.  When I shorted that resistor, it fixed everything.  

I found a CE-2 schem online and it shows a 33 ohm resistor (R52) coming off V+.  I don't know how accurate this schem is, but I know it's a really small value resistor.

http://www1.korksoft.com/~schem/veffects/bossce2.pdf

Good luck!
Ben

You might check to see if this small value resistor is on all ACA Boss pedals as well.  This could fix all ACA pedals to PSA.  

Ben

On the CE-2/B schematics I have there is a 390R resistor (R53 on the schem) and a 1S2473 diode (D5) in between the adapter negative and the input jack ring terminal.  The exact value of the resistor and diode type seems to vary from pedal to pedal.  I have schems for the DM-2/3, SD-1, SG-1, OD-1, NF-1, TW-1, GE-7, BF-2 and OC-2 and 470R seems most common.

Quote from: seekup1982The ACA power supplies are dirtier power but they run at a slightly higher voltage than 9v like batteries.  Boss added a little resistor here to lower the voltage when using a power supply.  If you simply short this resistor out, then a regular 9v power supply will work.  

I found this out by using a Visual Sound 1Spot on my Japan CE-2 and the light was lower than when I used a battery with it.  Also, it didn't sound as good.  When I shorted that resistor, it fixed everything.  

I found a CE-2 schem online and it shows a 33 ohm resistor (R52) coming off V+.  I don't know how accurate this schem is, but I know it's a really small value resistor.

http://www1.korksoft.com/~schem/veffects/bossce2.pdf

Good luck!
Ben

You need to remove the zener-diode (d7)  NOT r52 (which is part of the RC smoothing- filter with c30).

I would remove d5, and r53 and jumper the holes left. Remove the zener (d7) and replace it with d5 (for reverse-polarity protection), and change c30 to 16v or 25v --surprising to see a 10v cap there  (as Mark H notes

-Eric

The ACA to PSA conversion is on the link I put on my first post on this thread. Works fine.

Quote from: Fret WireThe ACA to PSA conversion is on the link I put on my first post on the thread.

So it is
Nice to see someone agrees with me.
C30 in my ce2 is a 10v --as in the scheme. I'd change that if you were going to run higher voltage, though you could have bias probs. if you use a battery (as you and Mark noted). I run my with a 9v regulated supply, and have no headroom problems. It should be noted that the 4558 has better noise and distortion specs than the BBD chip (it's actually a pretty good opamp), and I doubt you'd gain much by subbing it. YMMV

-Eric

You're right Eric, I'll probably replace c30 with a 16v cap, just for peace of mind. I'm gonna stay with 9v, so one trimpot setting will work for both batteries and power supplies. JPL Audio was nice enough to send a calibration sheet for scope set-up (layout and schematic). I'm gonna give the op amps a try, can't hurt. Interested in seeing what the NE5532 does. It quiets a TS-9 nicely. Gonna do the basic high-fi mod, and see how it sounds.