MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

Help sussing a non-working Roland SH-3A VCO?
MUFF WIGGLER Forum Index -> Music Tech DIY  
Author Help sussing a non-working Roland SH-3A VCO?
rjd2rjd2
[/img]

Hi all, I'm wondering if anyone may have any help; I'm not getting an output of the VCO on this machine I'm restoring. power supply looks right, and the whole PCB board has +/-15V as it should.

The IC labelled "upA41C", or TR129, is a transistor pair in an 8 pin DIP housing. I socketed this to provide easier diagnosis. I see no output there; that should be the VCO core, right? I'm of course not seeing anything at the output pin 4 of the "pack 8", which i'm guessing is the waveshaper section.

If I pull upA41C/TR129, I see +15V at the emitter pins on the socket; when in place, I read about .5V. does this sound right? I tried swapping this with a pair of 2n3906's, and i got the exact same behavior.

Thanks for any help; I'm not exactly sure how I should see that transistor pair behaving.
ricko
Yes, pin 4 of pack 8 is the VCO out. It is a high frequency signal, which gets divided down (It will be between 4 and 16 times the frequency of the 8' mix. i.e.x2 for 4', x2 for 2', x2 for L transpose, x2 for M transpose )

The thing that normally blows in the 3A is what pin 4 connects to on the next board: the LM3216 divider. (These are a little rare now, but could be simulated with counter chips.)

I think TR129 is the exponential converter. Current out?

The VCO core is everything in pack 8 plus the timing cap C141, is it? It uses a thyristor, which is not seen in DIY VCOs much. (I think it is like a "PUT relaxation oscillator".) Perhaps the LHS of the 0.47 cap on pack 8 is a good place to probe?
rjd2rjd2
aww man, that is music to my ears. thanks ricko!!! i'm gonna pull that lm3216, socket it, and then measure pin 4 of pack 8.

so, do you know how one approximates an LM3216? i've got a gang of components around here... i'll do some googling too. thanks!
ricko
Take care with the LM3216, because it is CMOS and fryable.

The 3A has a range of 9.5 octaves. But if you look at the design, the VCO only needs to track the 3.5 octaves of the keyboard: everything else is done by dividing down. ARP used the same trick with the Pro-Soloist. I think this is why Roland gets away with using the thyristor, which is not supposed to give good tracking outside a narrow range.

The other trick Roland had is in the resistor ladder in the keyboard: to avoid needing impossibly high precision resistors, they also allow resistors between other intervals (e.g. a high value resistor between some thirds or fifths). This lets them correct for accumulated error along the keyboard and even for the thyristor limitations, so the accuracy of the keyboard 8ves are pretty good, without the builders having to worry too much about semi-tone exactness.

If it turns out the problem is the CMOS divider, and you want a hint on rolling your own LM3216 check out
http://www.armory.com/~rstevew/Public/SoundSynth/TopOctave/topdividers .html#Fig15
rjd2rjd2
thanks ricko-i've never built a voltage divider network out of transistors; you don't by chance have any hunches as to what resistor values i'd want to use if i needed to do divisions by 2, and was using this circuit?

https://www.google.com/search?q=transistor+as+voltage+divider&tbm=isch &source=iu&ictx=1&fir=YOHffxv1VqvzwM%253A%252CnACU_vBaft-HeM%252C_&usg =__j0CdRMyscrnLf2-TOojCAbFxElk%3D&sa=X&ved=0ahUKEwio7fHJpabZAhXsmOAKHc wCDlYQ9QEIfjAO#imgrc=FfSHwoATHd0O-M:
MUFF WIGGLER Forum Index -> Music Tech DIY  
Page 1 of 1
Powered by phpBB © phpBB Group