Updated SSI2130 VCO Datasheet available...

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: Kent, luketeaford, Joe., lisa

Post Reply
DozenCrows
Common Wiggler
Posts: 111
Joined: Sun Jun 04, 2017 5:28 am

Re: Updated SSI2130 VCO Datasheet available...

Post by DozenCrows » Sat Nov 07, 2020 12:13 pm

I dialed my board down to about 7Hz, and am seeing the same sort of distortion on the sine begin to appear

Maybe something to do with the size of cap used between pins 1 and 2?

Ayeh
Learning to Wiggle
Posts: 18
Joined: Sat Nov 07, 2020 11:33 am

Re: Updated SSI2130 VCO Datasheet available...

Post by Ayeh » Sat Nov 07, 2020 12:25 pm

DozenCrows wrote:
Sat Nov 07, 2020 12:08 pm
Ayeh wrote:
Sat Nov 07, 2020 11:47 am
The only thing that bothers me is the sine shaper at low freq, it gets distorted as you can see in the pics
How low is low frequency here? My guess from your oscilloscope output is LFO range, around 1.5Hz?
Ok seems increasing the capacitor from sin+ will solve the issue, it seems is filtering low freq :party:
Attachments
IMG_6392.jpg

DozenCrows
Common Wiggler
Posts: 111
Joined: Sun Jun 04, 2017 5:28 am

Re: Updated SSI2130 VCO Datasheet available...

Post by DozenCrows » Sat Nov 07, 2020 12:29 pm

I checked the triangle wave also at that frequency, and it might be very slightly curved compared to higher frequencies. But nothing like the sine.

User avatar
neil.johnson
Super Deluxe Wiggler
Posts: 1571
Joined: Thu Apr 08, 2010 3:51 pm
Location: Cambridge, UK
Contact:

Re: Updated SSI2130 VCO Datasheet available...

Post by neil.johnson » Sat Nov 07, 2020 2:51 pm

Ayeh wrote:
Sat Nov 07, 2020 11:47 am
Hello peps,
I´m very pleased to write my first post in this forum about this incredible IC, :yay:
I´ve made a PCB following the schematics from the website and I´ve managed to solder this tiny QFN with a soldering iron, everything is working as expected, V/oct tracking and stability is very good after leaving the VCO on for a couple of hours, and TZ sounds pretty cool.
The only thing that bothers me is the sine shaper at low freq, it gets distorted as you can see in the pics, I did exactly what datasheet suggest. Is this normal behavior from the sinewave?? :despair:
Yes, at very low frequencies the series capacitor forms a high-pass filter with the resistors so you will get distortion as you are seeing. If you need to run a infrasonic frequencies then you would do better with the DC-coupled solution: short out the capacitor, then add a 12k7 resistor from VREF to the other input of the sine converter.

Neil
Random ramblings :: http://www.njohnson.co.uk
SSI parts and more :: https://www.cesyg.co.uk

User avatar
neil.johnson
Super Deluxe Wiggler
Posts: 1571
Joined: Thu Apr 08, 2010 3:51 pm
Location: Cambridge, UK
Contact:

Re: Updated SSI2130 VCO Datasheet available...

Post by neil.johnson » Sat Nov 07, 2020 2:57 pm

DozenCrows wrote:
Sat Nov 07, 2020 12:06 pm
One thing not explicitly mentioned in the datasheet is the expected signal amplitude from the mixer when a channel is set to unity gain. Unlike the tri, saw and pulse outputs, the mixer output is bipolar. When putting the indicated 100uA into one of the corresponding mixer channels, I get a ~1.6Vpp waveform on the output. The mix output is connected to GND via a 5K resistor as indicated in the datasheet.
It's a bit more complicated than that, and we're trying to word it in a way that is easy to understand. There are two ways of connecting the mixer output. The output is a current, and a simple resistor will convert that to a voltage. BUT.... its amplitude is limited to about +/- 3V. So we put in attenuation to allow mixing of several inputs without clipping. If you want higher amplitude signals then the better solution is to feed the mix out into a transimpedance amplifier (think inverting op-amp without the input resistor). Then you can scale the output by the feedback resistor, and the amplitude is now limited by the op-amp.

Neil
Random ramblings :: http://www.njohnson.co.uk
SSI parts and more :: https://www.cesyg.co.uk

User avatar
neil.johnson
Super Deluxe Wiggler
Posts: 1571
Joined: Thu Apr 08, 2010 3:51 pm
Location: Cambridge, UK
Contact:

Re: Updated SSI2130 VCO Datasheet available...

Post by neil.johnson » Sat Nov 07, 2020 4:06 pm

Important announcement regarding the SSI2130: DO NOT CONNECT THE THERMAL PAD TO GROUND

I've run some tests and, while it won't make things go bang, it must not be grounded as this will likely affect the tuning performance.

The next revision of the datasheet will clearly state this.

Neil
Random ramblings :: http://www.njohnson.co.uk
SSI parts and more :: https://www.cesyg.co.uk

User avatar
Stab Frenzy
Ultra Wiggler
Posts: 865
Joined: Wed Mar 03, 2010 5:46 am
Location: Melbourne

Re: Updated SSI2130 VCO Datasheet available...

Post by Stab Frenzy » Sat Nov 07, 2020 6:52 pm

Thanks for all your info on these Neil, and great work with the datasheets. I’ve got a couple of questions regarding the internal mixer that I’m sure I’ll find out soon enough but knowing before I send my boards off might save me a revision.

1. How do the mixer level inputs react to getting more current than required for unity gain? Do they top out at unity or attempt to amplify and distort the waveforms?

2. How do the same inputs react to being asked to source current, ie if they were connected to a bipolar LFO? I would assume the signal will be fully attenuated for the negative part of the cycle but don’t want anything to go pop.

3. This might be answered by your comment regarding the mix output being bipolar, but I was wondering if there were any quirks regarding mixing the unipolar internal waves forms with a bipolar signal on the aux inputs, seeing as they would be biased 1/VRef apart. Can I safely just not consider this or could there be issues caused by asymmetric clip points or similar?

Thanks again, I’m very excited to see what I can do with this.

DozenCrows
Common Wiggler
Posts: 111
Joined: Sun Jun 04, 2017 5:28 am

Re: Updated SSI2130 VCO Datasheet available...

Post by DozenCrows » Sun Nov 08, 2020 4:36 am

Stab Frenzy wrote:
Sat Nov 07, 2020 6:52 pm
3. This might be answered by your comment regarding the mix output being bipolar, but I was wondering if there were any quirks regarding mixing the unipolar internal waves forms with a bipolar signal on the aux inputs, seeing as they would be biased 1/VRef apart. Can I safely just not consider this or could there be issues caused by asymmetric clip points or similar?
I patched the sine output to the aux inputs to try this out, and it worked as expected. The internal waveforms are converted to bipolar and then mixed with the aux inputs.

Ayeh
Learning to Wiggle
Posts: 18
Joined: Sat Nov 07, 2020 11:33 am

Re: Updated SSI2130 VCO Datasheet available...

Post by Ayeh » Sun Nov 08, 2020 5:47 am

neil.johnson wrote:
Sat Nov 07, 2020 2:51 pm
Ayeh wrote:
Sat Nov 07, 2020 11:47 am
Hello peps,
I´m very pleased to write my first post in this forum about this incredible IC, :yay:
I´ve made a PCB following the schematics from the website and I´ve managed to solder this tiny QFN with a soldering iron, everything is working as expected, V/oct tracking and stability is very good after leaving the VCO on for a couple of hours, and TZ sounds pretty cool.
The only thing that bothers me is the sine shaper at low freq, it gets distorted as you can see in the pics, I did exactly what datasheet suggest. Is this normal behavior from the sinewave?? :despair:
Yes, at very low frequencies the series capacitor forms a high-pass filter with the resistors so you will get distortion as you are seeing. If you need to run a infrasonic frequencies then you would do better with the DC-coupled solution: short out the capacitor, then add a 12k7 resistor from VREF to the other input of the sine converter.

Neil
Great I will try that!

Thanks Neil! :sb:

User avatar
BugBrand
Knowledge of Bugs
Posts: 7545
Joined: Tue Jul 15, 2008 11:59 am

Re: Updated SSI2130 VCO Datasheet available...

Post by BugBrand » Sun Nov 08, 2020 6:13 am

Glad to hear the sine tip too because I'm def keen to try the 2130 with larger timing caps for LFO usage.
Have you tried such lo-freq approaches, Neil? (might well have been mentioned before...!)

User avatar
neil.johnson
Super Deluxe Wiggler
Posts: 1571
Joined: Thu Apr 08, 2010 3:51 pm
Location: Cambridge, UK
Contact:

Re: Updated SSI2130 VCO Datasheet available...

Post by neil.johnson » Sun Nov 08, 2020 12:23 pm

Stab Frenzy wrote:
Sat Nov 07, 2020 6:52 pm
Thanks for all your info on these Neil, and great work with the datasheets. I’ve got a couple of questions regarding the internal mixer that I’m sure I’ll find out soon enough but knowing before I send my boards off might save me a revision.

1. How do the mixer level inputs react to getting more current than required for unity gain? Do they top out at unity or attempt to amplify and distort the waveforms?

2. How do the same inputs react to being asked to source current, ie if they were connected to a bipolar LFO? I would assume the signal will be fully attenuated for the negative part of the cycle but don’t want anything to go pop.

3. This might be answered by your comment regarding the mix output being bipolar, but I was wondering if there were any quirks regarding mixing the unipolar internal waves forms with a bipolar signal on the aux inputs, seeing as they would be biased 1/VRef apart. Can I safely just not consider this or could there be issues caused by asymmetric clip points or similar?

Thanks again, I’m very excited to see what I can do with this.
1. I am removing the term "unity gain" removed from the data sheet as it depends on many factors, and in the example circuits shown is incorrect. The correct way to look at the mixer is the following expression for each channel:

Iout = Vin x Ictrl x Kgain

Kgain depends on which channel: for internal channels it is 1.28, and 0.64 for the external channels. Also bear in mind the maximum voltage on the MIX OUT pin is +/- 3V and you can see that unity gain is too narrow a term to use. The resistor on the output is fine for small circuits, but for anything decent it is better to connect the MIX OUT pin to an opamp and get much better signal levels.

2. The control inputs should be fine with negative control voltages - they are already limited by 100k resistors. I'll try just to be sure :tu:

3. The three internal signals are shifted by Vref/2 and scaled (see above about Kgain).

Neil
Random ramblings :: http://www.njohnson.co.uk
SSI parts and more :: https://www.cesyg.co.uk

User avatar
neil.johnson
Super Deluxe Wiggler
Posts: 1571
Joined: Thu Apr 08, 2010 3:51 pm
Location: Cambridge, UK
Contact:

Re: Updated SSI2130 VCO Datasheet available...

Post by neil.johnson » Sun Nov 08, 2020 12:27 pm

BugBrand wrote:
Sun Nov 08, 2020 6:13 am
Glad to hear the sine tip too because I'm def keen to try the 2130 with larger timing caps for LFO usage.
Have you tried such lo-freq approaches, Neil? (might well have been mentioned before...!)
The sine converter is fully static, so there is no limit to how slow you can drive it. You can put DC in it if you want. Another fun thing you can do with it is drive it from the sawtooth output, then you get a half-sine wave.

Neil
Random ramblings :: http://www.njohnson.co.uk
SSI parts and more :: https://www.cesyg.co.uk

User avatar
neil.johnson
Super Deluxe Wiggler
Posts: 1571
Joined: Thu Apr 08, 2010 3:51 pm
Location: Cambridge, UK
Contact:

Re: Updated SSI2130 VCO Datasheet available...

Post by neil.johnson » Sun Nov 08, 2020 12:30 pm

DozenCrows wrote:
Sun Nov 08, 2020 4:36 am
Stab Frenzy wrote:
Sat Nov 07, 2020 6:52 pm
3. This might be answered by your comment regarding the mix output being bipolar, but I was wondering if there were any quirks regarding mixing the unipolar internal waves forms with a bipolar signal on the aux inputs, seeing as they would be biased 1/VRef apart. Can I safely just not consider this or could there be issues caused by asymmetric clip points or similar?
I patched the sine output to the aux inputs to try this out, and it worked as expected. The internal waveforms are converted to bipolar and then mixed with the aux inputs.
The nice thing about current outputs is that is is very easy to mix multiple outputs - just one opamp to sum all the MIX OUT currents.

Neil
Random ramblings :: http://www.njohnson.co.uk
SSI parts and more :: https://www.cesyg.co.uk

User avatar
BugBrand
Knowledge of Bugs
Posts: 7545
Joined: Tue Jul 15, 2008 11:59 am

Re: Updated SSI2130 VCO Datasheet available...

Post by BugBrand » Sun Nov 08, 2020 12:40 pm

neil.johnson wrote:
Sun Nov 08, 2020 12:27 pm
BugBrand wrote:
Sun Nov 08, 2020 6:13 am
Glad to hear the sine tip too because I'm def keen to try the 2130 with larger timing caps for LFO usage.
Have you tried such lo-freq approaches, Neil? (might well have been mentioned before...!)
The sine converter is fully static, so there is no limit to how slow you can drive it. You can put DC in it if you want. Another fun thing you can do with it is drive it from the sawtooth output, then you get a half-sine wave.

Neil
Ah, the question was more for the VCO operation rather than the sine section!
[I have a feeling I asked the question before so should check back!]

User avatar
neil.johnson
Super Deluxe Wiggler
Posts: 1571
Joined: Thu Apr 08, 2010 3:51 pm
Location: Cambridge, UK
Contact:

Re: Updated SSI2130 VCO Datasheet available...

Post by neil.johnson » Sun Nov 08, 2020 2:35 pm

Ah right, the lower limit will be determined by the size and quality of the timing capacitor. At low charge currents any leakage will cause problems - asymmetry in the triangle shape, oscillator failing to oscillate, etc. Also, cost. Large low-leakage capacitors are very expensive. What's the largest polystyrene capacitor you can get? Mica? Teflon/PTFE? In the data sheet we use 3.9nF as that covers the audio range quite nicely, and will go down to a few Hz. You could go to 100nF for LFO duties. Although I'm sure if budget was no problem then you could "go large"....
https://www.digikey.com/en/products/det ... 10/7674872

Neil
Random ramblings :: http://www.njohnson.co.uk
SSI parts and more :: https://www.cesyg.co.uk

User avatar
BugBrand
Knowledge of Bugs
Posts: 7545
Joined: Tue Jul 15, 2008 11:59 am

Re: Updated SSI2130 VCO Datasheet available...

Post by BugBrand » Sun Nov 08, 2020 4:17 pm

Thanks - was hoping that would be the case - just wondered if there were any gotchas.

Pah, that cap is only +/-10% - what must a 1 or 2.5% cost?!!

User avatar
neil.johnson
Super Deluxe Wiggler
Posts: 1571
Joined: Thu Apr 08, 2010 3:51 pm
Location: Cambridge, UK
Contact:

Re: Updated SSI2130 VCO Datasheet available...

Post by neil.johnson » Sun Nov 08, 2020 4:52 pm

BugBrand wrote:
Sun Nov 08, 2020 4:17 pm
Thanks - was hoping that would be the case - just wondered if there were any gotchas.

Pah, that cap is only +/-10% - what must a 1 or 2.5% cost?!!
I'm guessing, but you probably don't want to know :eek:

Neil
Random ramblings :: http://www.njohnson.co.uk
SSI parts and more :: https://www.cesyg.co.uk

User avatar
KSS
Super Deluxe Wiggler
Posts: 4633
Joined: Mon Jan 25, 2016 7:28 am

Re: Updated SSI2130 VCO Datasheet available...

Post by KSS » Sun Nov 08, 2020 6:06 pm

But you wouldn't want to use that cap anyways. Not when a PP or PPS cap of that value or larger is widely available at a decent price in tight tolerances.
There's also this thing called parallel caps.

User avatar
neil.johnson
Super Deluxe Wiggler
Posts: 1571
Joined: Thu Apr 08, 2010 3:51 pm
Location: Cambridge, UK
Contact:

Re: Updated SSI2130 VCO Datasheet available...

Post by neil.johnson » Mon Nov 09, 2020 2:49 am

neil.johnson wrote:
Sun Nov 08, 2020 4:52 pm
BugBrand wrote:
Sun Nov 08, 2020 4:17 pm
Thanks - was hoping that would be the case - just wondered if there were any gotchas.

Pah, that cap is only +/-10% - what must a 1 or 2.5% cost?!!
I'm guessing, but you probably don't want to know :eek:

Neil
In a practical circuit you could try film caps, such as polypropylene. And if you need more capacitance you could always run a couple in parallel.

Cheers,
Neil
Random ramblings :: http://www.njohnson.co.uk
SSI parts and more :: https://www.cesyg.co.uk

User avatar
BugBrand
Knowledge of Bugs
Posts: 7545
Joined: Tue Jul 15, 2008 11:59 am

Re: Updated SSI2130 VCO Datasheet available...

Post by BugBrand » Mon Nov 09, 2020 4:28 am

Somehow I often forget about parallel caps (same note-to-self in FShift experiments at present!).
Yep, well used to swapping a VCO timing cap for larger to bring LFO realms - looking forward very much now to my test board arriving!

User avatar
Stab Frenzy
Ultra Wiggler
Posts: 865
Joined: Wed Mar 03, 2010 5:46 am
Location: Melbourne

Re: Updated SSI2130 VCO Datasheet available...

Post by Stab Frenzy » Tue Nov 10, 2020 12:00 am

neil.johnson wrote:
Sun Nov 08, 2020 12:23 pm
Stab Frenzy wrote:
Sat Nov 07, 2020 6:52 pm
Thanks for all your info on these Neil, and great work with the datasheets. I’ve got a couple of questions regarding the internal mixer that I’m sure I’ll find out soon enough but knowing before I send my boards off might save me a revision.

1. How do the mixer level inputs react to getting more current than required for unity gain? Do they top out at unity or attempt to amplify and distort the waveforms?

2. How do the same inputs react to being asked to source current, ie if they were connected to a bipolar LFO? I would assume the signal will be fully attenuated for the negative part of the cycle but don’t want anything to go pop.

3. This might be answered by your comment regarding the mix output being bipolar, but I was wondering if there were any quirks regarding mixing the unipolar internal waves forms with a bipolar signal on the aux inputs, seeing as they would be biased 1/VRef apart. Can I safely just not consider this or could there be issues caused by asymmetric clip points or similar?

Thanks again, I’m very excited to see what I can do with this.
1. I am removing the term "unity gain" removed from the data sheet as it depends on many factors, and in the example circuits shown is incorrect. The correct way to look at the mixer is the following expression for each channel:

Iout = Vin x Ictrl x Kgain

Kgain depends on which channel: for internal channels it is 1.28, and 0.64 for the external channels. Also bear in mind the maximum voltage on the MIX OUT pin is +/- 3V and you can see that unity gain is too narrow a term to use. The resistor on the output is fine for small circuits, but for anything decent it is better to connect the MIX OUT pin to an opamp and get much better signal levels.

2. The control inputs should be fine with negative control voltages - they are already limited by 100k resistors. I'll try just to be sure :tu:

3. The three internal signals are shifted by Vref/2 and scaled (see above about Kgain).

Neil
Excellent, that is all great information, thank you very much.

DozenCrows
Common Wiggler
Posts: 111
Joined: Sun Jun 04, 2017 5:28 am

Re: Updated SSI2130 VCO Datasheet available...

Post by DozenCrows » Sat Nov 14, 2020 7:50 am

Through-zero FM is so slick with this IC! I spent ages trying to get glitch-free TZFM using a 3340, but literally ten minutes or so with the 2130 even without the datasheet circuit.
DS1Z_QuickPrint30_crop.png

DozenCrows
Common Wiggler
Posts: 111
Joined: Sun Jun 04, 2017 5:28 am

Re: Updated SSI2130 VCO Datasheet available...

Post by DozenCrows » Sat Nov 14, 2020 11:53 am

Maybe spoke too soon...

I think I'm getting a slight drop in the 2130's pitch as I start to increase the modulation. This is manifesting as a beating in the output from relatively low levels of modulation (i.e. way before the zero point) even when the modulator and carrier are tuned to the same pitch (and the modulator is definitely not dropping in pitch). By detuning the modulator slightly with respect to the 2130 I can reduce the beating.

Probably an issue with the circuit feeding pin 21 - it's a precision full-wave rectifier using a TL072, cobbled together on a breadboard.

User avatar
neil.johnson
Super Deluxe Wiggler
Posts: 1571
Joined: Thu Apr 08, 2010 3:51 pm
Location: Cambridge, UK
Contact:

Re: Updated SSI2130 VCO Datasheet available...

Post by neil.johnson » Sat Nov 14, 2020 2:51 pm

How are you switching the TIME REVERSE pin? The recommended circuit uses a TL072 and an LM311 comparator, switching the TIME REVERSE and the modulating current at the same time.

Neil
Random ramblings :: http://www.njohnson.co.uk
SSI parts and more :: https://www.cesyg.co.uk

DozenCrows
Common Wiggler
Posts: 111
Joined: Sun Jun 04, 2017 5:28 am

Re: Updated SSI2130 VCO Datasheet available...

Post by DozenCrows » Sun Nov 15, 2020 4:42 am

neil.johnson wrote:
Sat Nov 14, 2020 2:51 pm
How are you switching the TIME REVERSE pin? The recommended circuit uses a TL072 and an LM311 comparator, switching the TIME REVERSE and the modulating current at the same time.
I'm using an LM311 in a similar way to the datasheet, driven by a signal from the first stage of the rectifier which does go negative. However I don't think this is part of the problem, as the shift in pitch occurs before FM depth is sufficient to cause the reverse to occur.

Here's the schematic for that part of the circuit. LINEAR_FM connects to pin 21, TRI_REVERSE to pin 18.
tzfm_2130.png
My current hypotheses (any or all may apply):
  • Rectifier is distorting/offsetting the modulator waveform
  • Original modulator waveform itself is asymmetric
The modulator is a 3340 with basic JFET sine shaper. All waveforms have been made bipolar and scaled to 10Vpp - but this isn't super-precise.

I get different effects on the FM with its different waveforms: for the same pitch, sine is the main case described above, square yields much more pronounced beating, and saw much more subtle beating.

I measured the average voltage on the output of U1B while increasing FM depth, expecting that to be pretty much constant - but it did change a bit, dropping from 2.5V with no FM to as low as 2.45V. This drop wasn't stable, but it might suggest a probable cause for the pitch change (unless it's a measurement error).

I'm going to check symmetry of the modulation signal, try different sources of modulation, try AC coupling the FM input, and more stuff when I think of it...

Post Reply

Return to “Music Tech DIY”