Yes I saw those. Very useful.
Here are some images.
I would mainly like to kow what readings to expect on the pins of the IC’s.
I may have faulty LM13700 chips.
Aside from the power pins, the voltages will depend on what signals you are putting in (and likely will be varying at audio frequency). The control inputs are current, not voltage; you can try measuring the voltage across the 10k resistors but it shouldn’t be possible for it to be too high so it probably wouldn’t tell you anything.
1 Like
Ok thank you.
But I’d like to know of the resistances I measured as in my post earlier are consistent with what would be expected.
Every time I power up the circuit with the lm13700 in place it blows the chip.
It appears your LM13700 socket is installed backwards; the notch should be toward the top in this view:
Of course the chip doesn’t care about the socket notch, you can put the chip in the other way around (with its notch up) and it’d be fine. Have you been putting the chip in with its notch up or down?
The 10k resistors on the control pins appear to be correct.
The pins you mention are the power pins for the chips, so it’s no surprise they’re the same resistances as each other and the power rails.
±12 V connect to the two ends of the cutoff pot, which is shown as 4.7k in the stripboard layout (though other values could be used), so that would account for 4.7k between the rails.
The wiper of the pot connects to a 100k resistor which connects to a 1.8k resistor to ground so that would account for a somewhat larger than 100k resistance from rails to ground. But there are other paths to ground as well which would lower the resistance some. Again, resistance measurements in circuit are rarely useful because of things like this. But 100k doesn’t surprise me.
None of these resistance values are particularly informative in diagnosing the problem other than to say the power rails evidently are not shorted to each other or to ground, and if they were you’d probably be seeing other (dramatic) symptoms.
While you have the chips out you might as well connect up the power supply and use a multimeter verify the +12 V really is +12 V and the -12 V really is -12 V at each of the chip power pins.
As for the possibly bad LM13700 chips, if you have a solderless breadboard it might be good to put together a minimal circuit from the datasheet, e.g.
and see if (1) the chip survives (2) it works. If you don’t have a breadboard maybe doing it on stripboard would be worthwhile. If it works it’d suggest there’s a problem with the VCF board, and if it doesn’t it’d mean it’s very likely the LM13700s are no good.
1 Like
Wow, Thank you so much for taking time to answer. Im quite new to electronics and am learning as fast as I can.
I shall try build that circuit of the typical application and let you know how it goes.
I may have more questions later.
Many thanks
I inserted the LM13700 chip with the notch toward the top.
So after building this filter twice it appears the problem was the LM13700 from eBay.
After purchasing Texas Instrument ones from R/S Components in the UK the circuit is working. (Apart from the LED’s don’t light up and the resonance is weak)
I got my money back on the cheap eBay ones.
As a warning to others, here is a picture of the dodgy chips.
The label reads 07JMLIRC3 LM13700N
1 Like
Looking at the logo — faq.tweakers.net/cme/iclogos appears to be down but it’s archived at https://web.archive.org/web/20220702085852/https://faq.tweakers.net/cme/iclogos/ and it shows the National Semiconductor logo like this:
or like this at Wikipedia:
and here it is on what I think is likely a legit chip, from https://octopart.com/adc1205ccj-1-national+semiconductor-491282:
If you compare that to your image you can see that on yours the logo is different, thinner and shaped a little differently, as well as being in outline instead of filled. The letters/numbers also look cruder. So I’m reasonably certain that’s a counterfeit.
1 Like
Well a lesson learned!
1 Like
Well no wonder the LED’s and the Resonance wasn’t working.
I forgot to add the 1.8K resistor.
I dont have any. Will a 2K resistor work?
And what effect would this have on the resonance been a higher resistance?
2k vs 1.8k won’t make much difference. But if you have anything in about the 15k to 22k range you could put that in parallel with the 2k to get close to 1.8k.
1 Like
Good idea.
I’ve actually dug out some 1.8K resistors that are ±5% I’ll test those.
Many thanks for your help.
I hope to be building your sequencer next!
I am wondering if my filter has a fake LM13700 , it’s never worked as expected.
Post a picture of it.
And where did you get it?
The first one I built works well as filter, resonance, etc… but it never had this craziness that one could expect.
The second one I built did work a lot better and I know I see why people find this filter quite aggressive (sound wise)
I used Kassutronics schematic for the second built in case you can spot something different (Kassutronics: KS-20 Filter)
Main differences I see:
Input resistor is 4.7k vs. 10k in the Schmitz version. That means the signal going into the filter is about twice as big (voltage divider is 4.7k over 220R instead of 10k over 220R).
Resistor to ground is 4.7k instead of the aforementioned 1.8k. That diminishes the maximum resonance feedback, by quite a bit.
That TI logo doesn’t look right. There is a pretty simple triangle/square vco in the data sheet that I use to check the chips. Checks both OTAs. You can look at the wave forms or hook up an LED to the square output to make a flasher for a quick check of basic operations.
1 Like