You are here

SH-2000: IC Intentionally Unsoldered

Customising, building or repairing your own gear? Need help with acoustic treatment or soundproofing? Ask away…

SH-2000: IC Intentionally Unsoldered

Postby zorgitron » Wed Apr 29, 2020 4:09 am

Hi all, I've been striving to bring an old Roland SH-2000 back to life.

It had some functionality, noise and VCF, but now I seemed to have made things worse and I humbly need to admit I don't know what I'm doing and consult some noble synth wisemen.

First, just all VCO signal went silent, but I could still hear the Noise with the "wind" preset and I could hear the VCF.

Then while poking around, the filter has gone silent as well. No noise either. And the LFO light isn't so blinky.

Anyway, while examining everything, nothing looked amiss except this old round type IC that has 2 legs just freely dangling in their solder holes.
Image
Image

Has anyone ever seen a case like this on a vintage synth? The holes have their solder pads covered with the green circuit board stuff. Also, checking the very blurry schematic, it looks like those legs aren't connected. (Such a damn blurry schematic though!)
Image
I hope this is the issue so I can just re-solder it and enjoy my SH 2000 again. But after killing 2 more sections of the synth, I'm kind of weary of proceeding.

Please help me out! Thanks for reading so far.
zorgitron
Posts: 2
Joined: Wed Apr 29, 2020 3:49 am

Re: SH-2000: IC Intentionally Unsoldered

Postby Hugh Robjohns » Wed Apr 29, 2020 10:25 am

zorgitron wrote:Anyway, while examining everything, nothing looked amiss except this old round type IC that has 2 legs just freely dangling in their solder holes.

That 'old round IC' is a CA3080 'Operational Transconductance Amplifier' or OTA dating from the early 70s in a round metal can rather than the flat 8-pin DIL chips that are more common these days.

Pins 1 and 8 of that OTA are not connected to anything in the chip -- as you have discovered -- and, since that board is obviously hand-soldered, I guess the person making the board up didn't want to waste time soldering pins that did nothing!

If you look at the solder pads, one has an inward pointing tail going nowhere -- that indicates the position of the marker tab on the case indicating pin 8 to ensure correct orientation. The pad next to it on the left is for pin 1. Neither do anything, so there's nothing at all to be concerned about over the missing solder!

I hope this is the issue so I can just re-solder it and enjoy my SH 2000 again.

Nope! You can solder them up if you really want to but it won't make any difference... and if you ever need to replace that OTA it will just make the job a lot harder!

But after killing 2 more sections of the synth, I'm kind of weary of proceeding.

I think I would be too.... There's nothing hugely complicated in the circuitry of the SH2000, but without a good understanding of electronics and the right test equipment you are more likely to wreck it than fix it!
User avatar
Hugh Robjohns
Moderator
Posts: 27115
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: SH-2000: IC Intentionally Unsoldered

Postby zorgitron » Thu Apr 30, 2020 12:21 pm

Thanks for answering my question.

I think I would be too.... There's nothing hugely complicated in the circuitry of the SH2000, but without a good understanding of electronics and the right test equipment you are more likely to wreck it than fix it!

I've wrecked many devices in my life! So I guess I'll just add this to the list and chuck the Roland in the bin.
Oh well.
zorgitron
Posts: 2
Joined: Wed Apr 29, 2020 3:49 am

Re: SH-2000: IC Intentionally Unsoldered

Postby Hugh Robjohns » Thu Apr 30, 2020 2:32 pm

I'm sure you'll find a buyer for it -- rather than binning it -- even if it is only partially working. Someone will be prepared to take it on as a project.

Alternatively, find an online study course on electronics, invest in some simple test equipment, and learn how to repair it properly! :D
User avatar
Hugh Robjohns
Moderator
Posts: 27115
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: SH-2000: IC Intentionally Unsoldered

Postby Adam Inglis » Fri May 01, 2020 10:45 am

Zorg, please don't bin it. There's some vintage hardware on this thing that is almost impossible to get, such as the flip tabs...
But I assume you said that out of frustration....
About a year ago I spent a great deal of time repairing mine. What makes troubleshooting difficult in this beast is the unusual architecture. It uses a matrix of diodes to set the values for the VCO waveforms and footages, the filter settings and the envelopes preset switches.
The presets have their various VCO, VCF and envelope values switched by FETs.

To cut to the chase, I fixed some absolutely bewildering problems simply by rehabilitating the .... wait for it... yes, the PSU.
The PSU should be your first port of call when troubleshooting any vintage synth!

The good news? There isn't any unobtainium in the circuitry so it should be fixable (by a tech who has carefully examined the schematic)
And more good news? It is a mind-blowingly-great sounding synthesiser!!
One of the most characterful in my studio (which is full of vintage synths). It sounds nothing like the other Roland synths of it's day, perhaps due to that ladder filter design they borrowed ;-) or maybe the very unusual VCO based on a unijunction transistor and a 555 chip.

Anyway, you have a hidden gem on your hands... good luck!

P.S. you can get a good quality PDF of the service notes from synfo.nl (perhaps leave a small donation)
P.P.S. more info from my investigations is here
http://mezzoauto.blogspot.com/2019/05/r ... -2000.html
User avatar
Adam Inglis
Regular
Posts: 422
Joined: Wed Sep 01, 2004 12:00 am
Location: Gold Coast Queensland Australia

Re: SH-2000: IC Intentionally Unsoldered

Postby Tomás Mulcahy » Fri May 01, 2020 12:48 pm

Just reading the blog Sam, very cool! Is the LM3216 divider on the VCO board or a matrix board?

Yes it is a very cool sounding synth. Very cutting, and ya, the Moog VCF. I regrettably had to part mine out as I acquired it without the two matrix boards (which by themselves are unobtainium unless you re-make them), which as Sam described are essential to the operation! They also have the actual envelopes, so even if I had put in pots to make the thing programmable, there were no envelopes. So now it is mostly parted out (still have to put the tab switches case and keyboard up on eBay).
I'm keeping the VCO and VCF boards to modularise them. Or at least- that's the plan :bouncy: The CV is little tricky apparently, it's not quite V/Oct.

Edit: Sam, I see you've figure out the scaling. Could you please share the circuit?
User avatar
Tomás Mulcahy
Frequent Poster
Posts: 1764
Joined: Wed Apr 25, 2001 12:00 am
Location: Cork, Ireland.

Re: SH-2000: IC Intentionally Unsoldered

Postby Adam Inglis » Sat May 02, 2020 8:12 am

Tomás Mulcahy wrote:Just reading the blog Sam, very cool! Is the LM3216 divider on the VCO board or a matrix board?

Tomás I assume you are addressing me?!!! Call me anything, I don't mind, just don't call me late for breakfast!
The divider chip is on a little daughterboard that plugs in to the main board (which you may well be missing). It's such a neat little system, easy to work on. The daughterboard provides the footages and the waveshapes, triangle, pulse, square - see page 8 of the SM.

Tomás Mulcahy wrote:I regrettably had to part mine out as I acquired it without the two matrix boards (which by themselves are unobtainium unless you re-make them).... They also have the actual envelopes, so even if I had put in pots to make the thing programmable, there were no envelopes.
Not exactly. The envelope board is another daughterboard. (Page 9 of the SM) The matrix switches in certain presets for Attack, Sustain and Release. I got round the limits of the presets by gaining external CV control of the volume, as described in the blog.

Tomás Mulcahy wrote:The CV is little tricky apparently, it's not quite V/Oct.
Edit: Sam, I see you've figure out the scaling. Could you please share the circuit?

The scaling is simply the usual 8 volt "Hz-per-octave" as used by Yamaha and Korg. I drive mine with a Kenton Pro Solo which gives me some aux CV s for the volume and the "Touch Effect" so I can bring in "growl" or "wah" or do pitch bend up or down (but not both!)

Here's an example of it in action on a new release
https://neinrecords.bandcamp.com/album/sorcery-2
The synth bass that opens the track and gradually gets more growly and rumbling and then starts pulsing in rhythm is the SH-2000 controlled in real time from the DAW. (Please excuse the flagrant self-promotion!!)
User avatar
Adam Inglis
Regular
Posts: 422
Joined: Wed Sep 01, 2004 12:00 am
Location: Gold Coast Queensland Australia

Re: SH-2000: IC Intentionally Unsoldered

Postby Tomás Mulcahy » Sat May 02, 2020 6:14 pm

Adam Inglis wrote:Tomás I assume you are addressing me?!!! Call me anything, I don't mind, just don't call me late for breakfast!
Oh I'm sorry! Got mixed up. Was worth it for the lols tho. Thank you :)

Love the track, especially the drums. I enjoyed reading back through your blog too.

I had thought my machine was only missing the two matrix boards, but obviously there were more. I just have the VCO and VCF boards. Which means the fabled waveshaper is not there, as you say, that's on a daughterboard (if I am understanding you).

Thanks for explaining the CV. Good to know! I'd heard different on the tinterwebs, remind me not to pay any attention to that. I'll see what I can figure out with the Expert Sleepers stuff. I sold my Pro Solo a long time ago (it used to drive 2 Yamaha CS5 and a CS10).

What I really want is the System 100M oscillator, played with one years ago and it was the biggest baddest thing I ever heard, patched straight to the VCA.
User avatar
Tomás Mulcahy
Frequent Poster
Posts: 1764
Joined: Wed Apr 25, 2001 12:00 am
Location: Cork, Ireland.

Re: SH-2000: IC Intentionally Unsoldered

Postby Adam Inglis » Sun May 03, 2020 12:38 am

Tomás Mulcahy wrote:Thanks for explaining the CV. Good to know! I'd heard different on the tinterwebs, remind me not to pay any attention to that.

The confusion comes from the schematic of the VCO. It notes the 'VCF key voltage' as 2 volts per octave, but this is only for filter tracking, not pitch control.

Tomás Mulcahy wrote:What I really want is the System 100M oscillator, played with one years ago and it was the biggest baddest thing I ever heard, patched straight to the VCA.
That came years later, 1980, and was more the typical saw-core VCO they used in their later analogues, such as the jupiter 8. That is volt-per-octave response. The new Berry clones should have you covered there!
User avatar
Adam Inglis
Regular
Posts: 422
Joined: Wed Sep 01, 2004 12:00 am
Location: Gold Coast Queensland Australia

Re: SH-2000: IC Intentionally Unsoldered

Postby Tomás Mulcahy » Mon May 04, 2020 3:07 pm

Thanks again! :thumbup:
User avatar
Tomás Mulcahy
Frequent Poster
Posts: 1764
Joined: Wed Apr 25, 2001 12:00 am
Location: Cork, Ireland.

Re: SH-2000: IC Intentionally Unsoldered

Postby RMCL » Sat May 09, 2020 10:48 am

Hi Zorgitron

The “green circuit stuff” that you refer to is solder resist. Although in this case the board has been hand soldered, it has been designed for wave soldering and as such the solder resist ensures that solder only goes on the locations on the board where a connection between the copper track and the device pin/leg is required. Solder resist minimises the amount of solder used, preventing bridging between adjacent tracks and as it remains in place after soldering process is complete prevents oxidation of the copper track.

Even although pins 1 & 8 of the device are not used, there should have been a hole in the solder resist on the PCB where the pins go through, so that if it had been wave soldered then a solder joint would have formed between the pin and the PCB track, rather than solder just ending up on the tips of pins 1 & 8 as they poke through the board.

With regards fault finding, even with just a multi-meter and a soldering iron some faults can be located and fixed by:

1. Visual inspection looking for physical damage, dry & cracked joints, cracked tracks (especially at the connectors or places where the board may flex) and swollen electrolytic capacitors (there is a strong case for just replacing all the electrolytic capacitors in this age of equipment anyway).
2. Checking supply voltages are ok
3. Point to point continuity checks (with power off) through connectors via board tracking to device pins.
4. Resistance checks (with power off) end to end & wiper to ends on the trim pots (do not adjust settings) and any other rotary or linear pots on the board.

Beyond this then a scope is required and with reference to the circuit diagram the waveforms throughout the board examined to see if they are as expected.
RMCL
Poster
Posts: 10
Joined: Sun Jan 27, 2019 9:31 pm