1222 Tuner VCO Module

My boards and panels (VCO and LFO) arrived a few days ago. I don’t know when I’ll have the spare money to populate and test them, but they are here and I can get on with the physical design using them as templates.

I’m pleased with the look and texture of the panels. That’s a good choice of material.

1 Like

Mine must have been here for nearly a month LOL… Just waiting on a final few components to turn up.

1 Like

Ok, first oscillator of three finally ready for calibration… wish me luck :stuck_out_tongue:

Pictured in my test bay with my LFO and VCF (both fully functional and ready to go)

Now I just need to find my little oscilliscope that I use for this kind of thing…

8 Likes

Good luck… Looks good.

I know exactly where my Old 20Mhz Leader Scope is as I keep stubbing my toe on it.

2 Likes

what kind of power supply do you use? i have these 3 modules incoming too.

1 Like

Well, in that test case I am running them off of a Make Noise Powered Skiff (you can see it in the background), but the final power solution I have for the proper case (that one is just a little short to fit modules comfortably in the bottom row, Doh!) is a pair of Freqency Central Routemaster boards, one per row. They are basically the same ones Sam uses, only they have an integrated bus board with 12 power headers each. I got them as complete kits from Thonk (where they also do kits for the FC Power and Microbus)

3 Likes

Oh, as a note, I built my #1112 Oscillator’s with ferrite beads as listed on the diagram, and they seem have enough resistance to cause PWM pitch drift when you change the PWM knob directly or apply a CV to the PWM input. This is a pretty common problem with the CEM3340’s as they are super sensitive to changes in current (on pin 3 if I remember right), but when I built my eurorack VCO Maximus boards, which are also 3340 designs, I used the same ferrites without issue so figured I would be ok. Unfortunatly this appears to not be the case. I will be taking them out and replacing them with jumpers to bypass the points like Sam does in his build video this weekend to see if it makes a difference. I also used the 47k resistor listed in the BOM which Sam swaps out for a 20k because he had the parts on hand, which might also be having an effect. They work a treat in the other modules mind you, although as with all things your milage may vary.

Has anyone used a AS3340-HYB in this circuit? Alfa claim that its a drop in replacement for an orginal CEM3340 only having fixed the problems with the PWM being sensitive and causing changes in frequency, but I haven’t tried it myself. If anyone has a good opinion of them I might order another set of PCB’s and give one a go to see what its like.

2 Likes

How much DC resistance do you measure on your ferrites? I would have expected close to zero at anything below tens of Megahertz.
Definitely let us know if removing them changes anything, as posted elsewhere on this site, ferrites beads should be mostly harmless/useless at audio frequencies.

1 Like

yes id be interested also. I have found that the power supply also effects the drift on Pwm I found it it was rather heavily loaded the Pwm would drift… Jon are you using the AS3340’s?? I have tried both there is a slight change with the cem3340, and I decided it was negligible, its much better with the AS3340 on the most extremes if you really listen there is a tiny bit, but I just rolled with it as it was right on the end of the Pam travel. a quick look at the Maximus design is looks similar enough the only difference I really see is the 15v.
I only have 1 of the VCO;s with ferrites on it and I haven’t noticed any ridiculous pitch drift, the shift is less than a couple of cents.
so this is rather interesting to hear what you are describing.
the ferrites I usesd were the type you put over a piece of resistor leg, not the ones you actually buy with legs

I’ll measure the resistance I am getting when I am actually back at home, but I also would expect it to be next to nothing (frankly I had just put em in because I had them to hand). They are the type that come supplied with their own legs. I picked them up from Soundtronics as a batch ages ago when I did a big order of stuff from them, but they gave very little in the way of information about the actual component other than ‘Ferrite Bead’, but they were the ones they recommended for their 5u kits.

As for the chip I am using the AS3340 in these ones, although I have used both in the past and found the difference in sound to be completely negligible (I have two of the Euro VCO Maximus boards from Fonitronik, one with a CEM and one with an Alfa so I could compare them). To be honest, on reflection, it might be down to the 47k resistor in R9 where you said you had spit the difference and just whacked in 20k’s.

Either way the thing sounds great, even with the drift. You are correct about it being more pronounced at the end of the PWM pot travel, I might just be being pedantic as its still pretty tuneful and its not making it go waaaay out of tune. Plus the detuning effect over the multiple oscillators (now I have the other two finished) is quite pleasant and makes the whole thing super thick sounding. One might even describe it as ‘phat’ :wink:

1 Like

I may look into a mod to reduce the pot travel. basically what I have noticed is on some designs you dont get the whole travel anyway, I think some designs have dealt with the drift by “clipping” the amount the Pulsw width can be adjusted. hence removing the extremes, which is where the drift happens. ill have a look and write a couple of suggested value changes to reduce the throw of the pulse width which will, if people want, eradicate the issue

1 Like

Does the PWM pot affect the pitch of the triangle and sawtooth outputs too?
I ask because the 3340 block diagram shows the pulse output as simply created with a comparator off of the sawtooth output, so it seems unlikely that the PWM would affect the frequency at all, but if it did, it would have to affect the other two outputs too, as there is really just one oscillator with three synchronous outputs.
Could it be that the drift is just a perceived illusion due to the weird harmonic content of narrow pulses?

1 Like

No, its definitely an actual shift in the output frequency. I had it hooked up to my DSO scope to check. Its a pretty well documented issue with the design of the chip itself.

1 Like

So it does also affect the other two outputs?

1 Like

I think so, I am pretty sure I was looking at the triangle output when I was calibrating it and thought to give the PWM a wiggle, although to be honest without getting my scope back out I can’t be 100% sure as I scoped all the outputs to make sure that the waveforms were correct.

1 Like

If the power supply is still suspected, you could try putting decoupling capacitors right on the 3340’s power pins (soldered directly to the socket pins) as the ones on the PCB are a bit far from the chip.
Actually C5 is on the wrong side of R6 if its purpose is to serve as a decoupling capacitor for the 3340.

If that fails to improve the situation due to a fundamental flaw in the 3340’s design, the solution might be to use an external comparator to create the pulse output.

1 Like

Oh, and for reference, I just checked the resistance on the ferrites and am getting about half an ohm when unpowered, so not quite zero.

1 Like

that could be enough!!! 1r resistors were making mine a bit too drifty at the ends!

1 Like

Mine measure about 0.1 ohm, did you factor out your meter’s leads?

1 Like

I am thinking the same, although I did just go through and recalibrate my reference voltage and tracking to make sure I hadn’t done something stupid. It was slightly out at between 4.010v to 4.040v across the oscillators and, now down to a stable 4.000v as far as my multimeter can tell. I mustn’t have left them on quite long enough to get to temperature when I did them first time round. I am now getting a drift of about 0.5% of the total freqency, so basically bugger all at most musical notes and only noticable on the highest PWM settings in the highest register of the selector knob. Granted if I play some stupidly high notes by passing a high CV in it’ll be more noticable but unless I want to torture the neighbours pets I think I will be fine :smiley:

1 Like