Roland RE-201: recapping power supply (reduce hum)

Hello everybody!
My Space Echo recently developed some excessive hum, so I'm about to recap the power supply. While I'm at it, I thought about increasing the values of the filter caps to further lower the hum level.
I'm thinking about C108/C109/C110 on the power supply board and C71 on the main PCB. (Maybe 2200uF/220uF/22uF?)

Has anybody done this in a Space Echo? Does this make sense? And do you have any tips on how to further lower the hum level?

Thanks!





Here is a link to the complete schematic:
http://manuals.fdiskc.com/flat/Roland%20RE-101%20&%20RE-201%20Service%20Manual.pdf

Just recap all the ageing PSU electros using the values stated on the original schematic with quality panasonic caps.
Then see if if the hum dissappears.... if not ,.......then you look elsewhere.
A scope always helps.

I don't know if you need to increase the filter caps much. The caps were probably selected to be oversized in the first place to account for the usual wear. If you do increase them, don't do it by more than 30% or so because a) inrush currents might stress something else and b) it might take noticeably longer for the unit to turn off.

The only exception might be the motor PS filter cap. I changed mine from 1000u to 2200u because that cap gets punished pretty hard and I thought there was a little excessive warble in the motor (although I'm not sure if it had an effect - there still is some warble at low speeds and my motor is pretty solid - it spins freely and there's no "play" in the shaft).

Other than PSU filter caps, the next best place would be bypass caps around the high gain parts. In particular the recovery amp:



I would replace at least the 100u caps because any hum at the recovery amp input is going to be amplified by a lot.

ECHO VOLUME BOOST MOD:

As long as I have that graphic up I want to share a mod that I think is a no-brainer for a Space Echo:

In my RE-101, the Echo Volume would not exceed the volume of the original signal. I don't know if that's just how it's supposed to work or if it's because of head wear or whatever but the delayed echo was always a little quieter by a few dB. I verified this using an audio analyzer.

However, there is a super easy mod that allows you to boost the level of the delayed sig to be louder than the non-delayed signal when the Echo Volume is at it's maximum position. I have mine modified so that when the Echo Volume is at 12 o'clock I get the unmodified behavior but at about 2 o'clock the levels are about the same and of course above that the delayed signal is actually is louder but about 4 dB.

This mod is an effect in itself. It really makes a difference with note doubling techniques in particular.

Anyway, the mod is very simple and it can be easily removed. All you have to do is piggy-back / tack on a 470 ohm resistor on top of the 270 ohm R67 in the above graphic. This changes that resistor from 270 ohms to 170 ohms (270||470) which equates to a 6dB boost in the recovery amp. This has no effect on DC bias of course because it's on the ground side of the bypass cap.

Note: Clean the existing resistor leads first with some flux. If they're covered in grime it may be hard to get the new resistor to stick.

Technically this does shift the HP frequency from 27 Hz to 42 Hz so you could change the bypass cap from 22u to 47u if you really wanted to be pedantic about it.But a little 6dB HP shelf is not necessarily bad.  I think I kept the 22u. But you should put a fresh cap there regardless.

Wow, thank you very much for the detailed response! 

Pretty much every Space Echo I heard had more or less hum on the output, that's why I thought about increasing caps. But as you said, inrush current when switching on is my main concern. The next value I have here is 2200uF, so I guess I'll better use the standard one...

And thanks for pointing out the ones in the recovery amp, I completely forgot about these!
I plan to replace all electrolytics anyway, but for now I'll just do the most important ones.

Interesting idea about increasing the gain of the playback head, although I don't really think I'd ever need a louder delay...
Does this also increase the SNR? My guess is that it doesn't, since most noise is also amplified...

You can reduce psu noise by adding more caps (and therefore capacitance) as well. 

Change the 1000u caps with same and then on the underneath add on another 220u or 330u or even 470u depending on what you have .

Personally I am of the opinion that increasing to 2200u would not be a problem .  from what I understand in rush current issues would only affect the diode bridge components, so it's pretty easy to compensate for the potential, change the diodes to something beefier.

Another way to reduce psu noise is to bypass the 1000u caps with c0g ceramics, 0.01 or even 0.1 caps.  Mouser sells these nice little blue ones.

andow said:

Pretty much every Space Echo I heard had more or less hum on the output, that's why I thought about increasing caps. But as you said, inrush current when switching on is my main concern. The next value I have here is 2200uF, so I guess I'll better use the standard one...

Click to expand...

If you really want to increase them to 2200u it probably would be fine even without beefing up diodes. However, I don't think it would help. Those caps are probably already larger than they need to be.

Hum is more likely to be caused by grounding issues. The Space Echo does not have balanced inputs / outputs. If you turn all of the levels down and unplug all of the inputs does the hum go away? If yes, that's a pretty strong indication of ground loop issues. Hum will be particularly bad if you connect the Space Echo to another unbalanced piece of gear that has it's own power supply because this will DC connect the two grounds together over the cable. Hum will be even worse still if the other piece of gear has a floating SMPS supply with no earth ground like just about every cheap piece of consumer electronics as in some plastic box with a wall wart PS.

Possible work-arounds / fixes for ground loop issues (not specific to Space Echos) are:

1) Connect it to only balanced gear or gear that is battery powered. For example, connect a guitar or guitar pedal running on battery (not wall supply so that guitar / pedal will have a floating ground) to the instrument input and then connect the output to something with a balanced input (this can be a transformer line input or electronically balanced) that does NOT connect the cable shield to pin 1 at whichever end. Most gear DOES connect the shield to pin 1 so you might have to make a cable for this by talking a TRS cable and removing one of the plugs and make it just a TS and connect wire 2 to T and wire 3 to S and leave wire 1 (the shield) unconnected. Then when connected to a balanced input the grounds will be isolated. Meaning the ground of the Space Echo will be connected to pin 3 of the remote balanced input instead of pin 1 which is ground / chassis. Like this:



You can do the same for the instrument input if the source is transformer balanced. I don' know if it would work with an electronically balanced output because wire 3 would be driving ground which might not be good for the output depending on the circuit. Another alternative is to get one of those transformer isolated 1/4 adapters. Or, for a top notch solution add a 10K line input transformer to the instrument input.

2) Add a 3rd prong earth ground power cable and try to connect all gear interconnected with the Space Echo to the same power strip.  My Space Echo came from Japan so I guess they didn't care about earth ground back then because the power cord was only a 2 prong deal. So I added a proper 3 prong earth ground cable and connected the earth ground directly to the chassis of the Space Echo. All of my music gear share power form the same power conditioner and power strips. This minimizes the difference between grounds of each piece of gear. It's also a good idea to partition high-power gear (like power amps or just large power hungry units) and low-power gear to use separate power strips b/c it's the high currents that cause differences in ground potential.

Incidentally, regarding Japanese units, they are wired for 100V. Fortunately mine had a power transformer with a second 120V primary tap (you can see this in the schematic) that was wired to a piece of tag board. So I was able to simply unsolder a wire and move it to a different terminal on the tag board and voila, I now have a 120V US unit.

andow said:

And thanks for pointing out the ones in the recovery amp, I completely forgot about these!
I plan to replace all electrolytics anyway, but for now I'll just do the most important ones.

Interesting idea about increasing the gain of the playback head, although I don't really think I'd ever need a louder delay...
Does this also increase the SNR? My guess is that it doesn't, since most noise is also amplified...

Click to expand...

Right. The noise will be amplified also so the SNR will not change. The louder echo is nice for note doubling effects like you hit a sequence of keys at even intervals and the echo lands in between so that it sounds like you're playing at double speed. Arpeggios on the guitar like Steve Vai in the Crossroads movie from the 80's did this (not to be confused with the Britney Spears movie!).

Beefing up the psu caps a bit definitely helped in my opinion. The echoes are more bold and clear now if that's your thing.