R*S Variable Slope capacitors

Discussing some incredible modules that don't quite fit into the other forum categories.

Moderators: Kent, luketeaford, Joe., lisa

Post Reply
mfaraday
Common Wiggler
Posts: 54
Joined: Sat Jan 31, 2015 9:49 am
Contact:

R*S Variable Slope capacitors

Post by mfaraday » Tue Oct 13, 2020 2:51 pm

hello!

i have a favor to ask: does anyone have a 4U random source variable slope filter (v1.02c) and could post a picture of the main pcb? mine does not seem to be acting quite right (very very low output compared to its VCFQ neighbor) and the pdf is a bit unclear about which caps go where.

below is a photo of mine at the moment. the random source build doc mentions omitting certain caps and has references to caps that don't quite line up with what's on the pcb. if anyone has a working version i could compare to i'd appreciate it very much!
IMG-0586.JPG
thanks!

User avatar
dksynth
Wiggling with Experience
Posts: 431
Joined: Wed Jul 01, 2015 12:56 pm

Re: R*S Variable Slope capacitors

Post by dksynth » Tue Oct 13, 2020 4:14 pm

could you please show the back of the PCB?

mfaraday
Common Wiggler
Posts: 54
Joined: Sat Jan 31, 2015 9:49 am
Contact:

Re: R*S Variable Slope capacitors

Post by mfaraday » Tue Oct 13, 2020 4:54 pm

flip side:
IMG-0604.JPG

User avatar
asymptote
Common Wiggler
Posts: 207
Joined: Mon Aug 31, 2015 11:16 pm
Location: Brighton, MA
Contact:

Re: R*S Variable Slope capacitors

Post by asymptote » Wed Oct 14, 2020 12:41 am

My VCFS is as yet unbuilt, but I've puzzled over this myself. The documentation and PCB are really unclear, especially with having both C & CB 5s, 6s, and 9s. The 4U build doc does mention that the omissions represent 'radical part choices,' and if you look at the euro version build doc there is no mention of omitting caps nor are there any unfilled locations on the pictured PCB. You might try just stuffing all the caps.

I also see that you're missing a cap at CB10. Given that there's no schematic for this circuit, I'm not sure how that would impact behavior. But I would address that first to see if it helps with the low output.

User avatar
dksynth
Wiggling with Experience
Posts: 431
Joined: Wed Jul 01, 2015 12:56 pm

Re: R*S Variable Slope capacitors

Post by dksynth » Wed Oct 14, 2020 10:06 am

C5, C6 and C9 all should be installed.

C5 and C6 are phase compensation caps for the integrators as used on most serge filters and likely less critical. C9 I'm not sure about but I tend to think that with the faster opamps driving things there that stability cap might be critical.

djs
Super Deluxe Wiggler
Posts: 1204
Joined: Wed Sep 01, 2010 8:36 am
Location: South Beloit, IL

Re: R*S Variable Slope capacitors

Post by djs » Wed Oct 14, 2020 11:36 am

I don't have one of these, but what's going on here?
2020-10-14 11_34_57-IMG-0604.JPG (JPEG Image, 2960 × 2960 pixels).png
"Noise is what the Earth is made of" - David Bowie

User avatar
asymptote
Common Wiggler
Posts: 207
Joined: Mon Aug 31, 2015 11:16 pm
Location: Brighton, MA
Contact:

Re: R*S Variable Slope capacitors

Post by asymptote » Wed Oct 14, 2020 11:46 am

dksynth wrote:
Wed Oct 14, 2020 10:06 am
C5, C6 and C9 all should be installed.

C5 and C6 are phase compensation caps for the integrators as used on most serge filters and likely less critical. C9 I'm not sure about but I tend to think that with the faster opamps driving things there that stability cap might be critical.
Do you know what’s up with the PCB having both C5, C6, C9 and CB5, CB6, CB9? The BOM doesn’t differentiate these positions and it specifies only 3 caps rather than 6. Interestingly, the picture in the euro build doc has all 6 installed, but with a mix of film and ceramic.

User avatar
B0bcat
Wiggling with Experience
Posts: 497
Joined: Sun Jun 10, 2018 9:07 pm

Re: R*S Variable Slope capacitors

Post by B0bcat » Wed Oct 14, 2020 11:50 am

asymptote wrote:
Wed Oct 14, 2020 11:46 am
dksynth wrote:
Wed Oct 14, 2020 10:06 am
C5, C6 and C9 all should be installed.

C5 and C6 are phase compensation caps for the integrators as used on most serge filters and likely less critical. C9 I'm not sure about but I tend to think that with the faster opamps driving things there that stability cap might be critical.
Do you know what’s up with the PCB having both C5, C6, C9 and CB5, CB6, CB9? The BOM doesn’t differentiate these positions and it specifies only 3 caps rather than 6. Interestingly, the picture in the euro build doc has all 6 installed, but with a mix of film and ceramic.
could somebody comment on what would be the "benefit" of having some of said caps ceramic? are they bypass caps where it doesn't matter?
WTB: MacBeth X-Series Single VCO / Dual Oscillator / Elements

FS(US): Malekko/Wiard Anti-Oscillator (Gargoyles), SSF Entity BD

User avatar
dksynth
Wiggling with Experience
Posts: 431
Joined: Wed Jul 01, 2015 12:56 pm

Re: R*S Variable Slope capacitors

Post by dksynth » Wed Oct 14, 2020 12:09 pm

asymptote wrote:
Wed Oct 14, 2020 11:46 am
Do you know what’s up with the PCB having both C5, C6, C9 and CB5, CB6, CB9? The BOM doesn’t differentiate these positions and it specifies only 3 caps rather than 6. Interestingly, the picture in the euro build doc has all 6 installed, but with a mix of film and ceramic.
The CB caps are all bypass caps, I think that's what the B is meant to stand for.
B0bcat wrote:
Wed Oct 14, 2020 11:50 am
could somebody comment on what would be the "benefit" of having some of said caps ceramic? are they bypass caps where it doesn't matter?
Bypass caps can always be ceramic as far as I know just pay SOME attention to what value they'll actually functionally have depending on DC and substrate. I would use high quality C0G/NP0 for the compensation caps or Mica as in the original if you feel like burning money. The C9 position is also fine as C0G/NP0 but the original I think just used standard ceramic. I don't currently have an original VCFS otherwise I would check for sure for you.

User avatar
B0bcat
Wiggling with Experience
Posts: 497
Joined: Sun Jun 10, 2018 9:07 pm

Re: R*S Variable Slope capacitors

Post by B0bcat » Wed Oct 14, 2020 12:15 pm

dksynth wrote:
Wed Oct 14, 2020 12:09 pm
asymptote wrote:
Wed Oct 14, 2020 11:46 am
Do you know what’s up with the PCB having both C5, C6, C9 and CB5, CB6, CB9? The BOM doesn’t differentiate these positions and it specifies only 3 caps rather than 6. Interestingly, the picture in the euro build doc has all 6 installed, but with a mix of film and ceramic.
The CB caps are all bypass caps, I think that's what the B is meant to stand for.
B0bcat wrote:
Wed Oct 14, 2020 11:50 am
could somebody comment on what would be the "benefit" of having some of said caps ceramic? are they bypass caps where it doesn't matter?
Bypass caps can always be ceramic as far as I know just pay SOME attention to what value they'll actually functionally have depending on DC and substrate. I would use high quality C0G/NP0 for the compensation caps or Mica as in the original if you feel like burning money. The C9 position is also fine as C0G/NP0 but the original I think just used standard ceramic. I don't currently have an original VCFS otherwise I would check for sure for you.
That is interesting. I’m only familiar with the Euro builds but it had never occurred to me to consider using Mica for the bypass caps, since everything annotated as bypass caps on the BOM is 10N or 100N caps. So I’m not sure we are talking about the same caps here :)
WTB: MacBeth X-Series Single VCO / Dual Oscillator / Elements

FS(US): Malekko/Wiard Anti-Oscillator (Gargoyles), SSF Entity BD

User avatar
dksynth
Wiggling with Experience
Posts: 431
Joined: Wed Jul 01, 2015 12:56 pm

Re: R*S Variable Slope capacitors

Post by dksynth » Wed Oct 14, 2020 12:20 pm

B0bcat wrote:
Wed Oct 14, 2020 12:15 pm
dksynth wrote:
Wed Oct 14, 2020 12:09 pm
asymptote wrote:
Wed Oct 14, 2020 11:46 am
Do you know what’s up with the PCB having both C5, C6, C9 and CB5, CB6, CB9? The BOM doesn’t differentiate these positions and it specifies only 3 caps rather than 6. Interestingly, the picture in the euro build doc has all 6 installed, but with a mix of film and ceramic.
The CB caps are all bypass caps, I think that's what the B is meant to stand for.
B0bcat wrote:
Wed Oct 14, 2020 11:50 am
could somebody comment on what would be the "benefit" of having some of said caps ceramic? are they bypass caps where it doesn't matter?
Bypass caps can always be ceramic as far as I know just pay SOME attention to what value they'll actually functionally have depending on DC and substrate. I would use high quality C0G/NP0 for the compensation caps or Mica as in the original if you feel like burning money. The C9 position is also fine as C0G/NP0 but the original I think just used standard ceramic. I don't currently have an original VCFS otherwise I would check for sure for you.
That is interesting. I’m only familiar with the Euro builds but it had never occurred to me to consider using Mica for the bypass caps, since everything annotated as bypass caps on the BOM is 10N or 100N caps. So I’m not sure we are talking about the same caps here :)
Important distinction highlighted in bold, my fault as I spread the info over two comments. Compensation caps are not the bypass caps. C5 and C6 appear to be phase compensation caps for the integrators.

User avatar
B0bcat
Wiggling with Experience
Posts: 497
Joined: Sun Jun 10, 2018 9:07 pm

Re: R*S Variable Slope capacitors

Post by B0bcat » Wed Oct 14, 2020 12:22 pm

that makes much more sense!! apologies for my miscomprehension
WTB: MacBeth X-Series Single VCO / Dual Oscillator / Elements

FS(US): Malekko/Wiard Anti-Oscillator (Gargoyles), SSF Entity BD

User avatar
asymptote
Common Wiggler
Posts: 207
Joined: Mon Aug 31, 2015 11:16 pm
Location: Brighton, MA
Contact:

Re: R*S Variable Slope capacitors

Post by asymptote » Wed Oct 14, 2020 12:25 pm

dksynth wrote:
Wed Oct 14, 2020 12:09 pm
asymptote wrote:
Wed Oct 14, 2020 11:46 am
Do you know what’s up with the PCB having both C5, C6, C9 and CB5, CB6, CB9? The BOM doesn’t differentiate these positions and it specifies only 3 caps rather than 6. Interestingly, the picture in the euro build doc has all 6 installed, but with a mix of film and ceramic.
The CB caps are all bypass caps, I think that's what the B is meant to stand for.
B0bcat wrote:
Wed Oct 14, 2020 11:50 am
could somebody comment on what would be the "benefit" of having some of said caps ceramic? are they bypass caps where it doesn't matter?
Bypass caps can always be ceramic as far as I know just pay SOME attention to what value they'll actually functionally have depending on DC and substrate. I would use high quality C0G/NP0 for the compensation caps or Mica as in the original if you feel like burning money. The C9 position is also fine as C0G/NP0 but the original I think just used standard ceramic. I don't currently have an original VCFS otherwise I would check for sure for you.
Awesome, very helpful info. Thanks!

mfaraday
Common Wiggler
Posts: 54
Joined: Sat Jan 31, 2015 9:49 am
Contact:

Re: R*S Variable Slope capacitors

Post by mfaraday » Wed Oct 14, 2020 12:29 pm

djs wrote:
Wed Oct 14, 2020 11:36 am
I don't have one of these, but what's going on here?
seems like some weird glare or something - looks fine on my board. my phone is really old and the camera is admittedly shit, lol

and thanks for all of this info! i'll just go ahead and install all of the caps and see if that works.

i also checked some of the resistor values that the BOM identifies as having to deal with output gain and they're correct per the BOM.... is the variable slope just known to have a lower output volume?

mfaraday
Common Wiggler
Posts: 54
Joined: Sat Jan 31, 2015 9:49 am
Contact:

Re: R*S Variable Slope capacitors

Post by mfaraday » Tue Oct 20, 2020 10:56 am

i'm wondering if the missing caps at C5, C6, and C9 are the only things to blame? i've installed all other caps, but before buying and installing C5/C6/C9 i'm curious if anything else might be going on. compared to the VCFQ i'm getting a much weaker and inverted signal. i'm not an expert, but per ralf's post here i don't think the absence of those caps would be causing this...?

is this just normal behavior for the VCFS? anything else i should be checking?

here are some traces (not the best scope in the world, but should suffice):

raw SAW out of PCO
IMG-0612.JPG
SAW -> VCFQ lo out (max gain + cutoff, min Q)
IMG-0615.JPG
SAW -> VCFS lo out (crossfader all the way to input, max cutoff, min Q + slope)
IMG-0617.JPG
thanks!

mfaraday
Common Wiggler
Posts: 54
Joined: Sat Jan 31, 2015 9:49 am
Contact:

Re: R*S Variable Slope capacitors

Post by mfaraday » Fri Nov 20, 2020 10:23 am

just an update on this - of course i messed up and used 330k resistors in place of 300 ohm resistors. always something silly like that!

i left C5, C6, and C9 unpopulated and seems to be working great!

one question i have is if it's typical for the variable slope output to be inverted? it seems pretty useful to me, but don't think i've seen anything written anywhere about that fact

User avatar
cygmu
Super Deluxe Wiggler
Posts: 1642
Joined: Mon Feb 15, 2016 3:33 pm
Location: UK

Re: R*S Variable Slope capacitors

Post by cygmu » Fri Nov 20, 2020 4:18 pm

mfaraday wrote:
Fri Nov 20, 2020 10:23 am
one question i have is if it's typical for the variable slope output to be inverted?
Yes, that’s how it is.

kevin0
Learning to Wiggle
Posts: 50
Joined: Fri May 26, 2017 6:11 am

Re: R*S Variable Slope capacitors

Post by kevin0 » Wed Nov 25, 2020 4:32 am

cygmu wrote:
Fri Nov 20, 2020 4:18 pm
mfaraday wrote:
Fri Nov 20, 2020 10:23 am
one question i have is if it's typical for the variable slope output to be inverted?
Yes, that’s how it is.
That's interesting. This must affect stereo patching of the filters on La Bestia 2 a lot, surely you have to flip the phase if you want to use it in parallel with lets say the VCFQ?

Is there some upside to the phase inversion that I'm not seeing?

User avatar
cygmu
Super Deluxe Wiggler
Posts: 1642
Joined: Mon Feb 15, 2016 3:33 pm
Location: UK

Re: R*S Variable Slope capacitors

Post by cygmu » Wed Nov 25, 2020 11:47 am

I would guess that the historical reason for this is that it saved an op-amp. These are designs from the 1970s when parts were not as cheap as they are now and PCB real-estate was also valuable.

Post Reply

Return to “Buchla, EMS & Serge”