VC LFO 2 FREQ NOT STABLE

[jansensouza]

Hi, I've assembled 4 of that VC LFO 2 and they are driving me nuts because all 4 havent a stable frequency. As I turn the modular on the frequency of those LFOs begins to rise for example from 100 Hz to 200Hz in 5 or 8 minutes, after a while it becomes stable but if you change the freq knob the situation returns and the freq begins to change. I would like to know if someone had some problem like that. Thanks in advance

[blue hell]

jansensouza

posting a link to the schematics would help people to find a possible cause.

[jansensouza]

sorry

http://yusynth.net/Modular/EN/LFO/VC-LFO2.html

thanks

[blue hell]

Hmm .. not direcly seeing anthing ..is it on both ranges?

[jansensouza]

Yes, and that's odd because after stabilizing you can change the range and it will be stable but if you change the potenciometer the frequence begins to vary for another period of 5 to 10 minutes, I am puzzled indeed.

[Hubaswift]

Odd behavior. I've built both the LFO-1 and LFO-2 and confirm that they should be stable from powerup. Have you checked the voltage rails to see if a small shift in voltage there could be throwing the off all your LFOs? If that is the case you'd also see a small drift in voltage on pin 6 of the 741. Also, just to be sure, are you using a 7555, not just a generic 555 timer? I don't know what kind of issues that would cause but I imagine it could warrant strange behavior.

[jansensouza]

I am using the 7555 and tomorrow I will change all capacitors as a test, the voltage regulators as long as I could see are ok, and yes there is a drift in voltage at pin 6 of the TL071. the question is strange because all 4 LFO have the same behavior. I should done some mistake and the same mistake in all 4...

[jansensouza]

another detail: it seems to me that the BC557b becames a little bit hot and when I touch it the frequence changes a lot ( dont need to mention that I've changed the transistor and nothing...)

[blue hell]

[jansensouza]

I am gonna try it
I'll let you know later

[jansensouza]

well I did the follow:
1- voltage lines check +15,022 V steady and -15,028V steady as a rock
2- removed the FET
3- changed the capacitors C1 and C3 from another type (but still polyester) (10nf each)
4- changed C5 that was 10nf for another of 100nf ( some prints shows 10nf and others 100nf)
5- checked the wiring of the range switch - OK

power up : the same crazy behavior

odd...

[blue hell]

You seem to have another schematic than the one you posted a link to?

I can not really map the linked one to the component designators / values you mention in your above post.

Anyway, that BC557 (Q1) going warm .. that should a) not happen and b) when it does it could explain the frequency drift.

It will get warm when too much current flows trough it, that was the idea of removing the FET and to check the wiring of the range switch .. apparently it is not that as those changes did not help ...

Other places excess current could flow into ... the 7555 (IC1), and IC2b (1/4 TL074) .. that is a bit harder to check ... and .. it would be odd that it would happen in all four builds ... R12 .. is that really 27k .. when it would be way smaller it might explain it too ...

One other thing you could check .. the supply currents, those have a mention in the schematic .. when your measured currents would be a lot larger .. that could be another indication of too much current flowing trough Q1.

[jansensouza]

I checked the r12 and it is 27K but on the meter reads 10 - 12 K in place.
I've already changed the 7555, TL071 and TL074

you said " One other thing you could check .. the supply currents, those have a mention in the schematic .. when your measured currents would be a lot larger .. that could be another indication of too much current flowing trough Q1."

where is the mention in the schematics?

[blue hell]

[blue hell]

[jansensouza]

The TL071 is in the component list as option to 741

http://yusynth.net/Modular/index_en.html

[jansensouza]

the current from the positive line : begins with 10,000 mA and after 10 min 11,800 mA and rising...

negative line : 8,246 mA steady

[blue hell]

Hmm ... I'm out of ideas now

This rising is an odd thing too tho.

[jansensouza]

I tried to change the TL071 for an 741 and nothing changed. I noticed that the transistor BC557b is getting real hot to touch, the resistor R13 was checked and is 820R and R6,7,9 and 11 are 100K 1%, r16 2,7k and R12 27k 1%
I removed the IC 3 (TL074) and put IC2's pin 7 to osciloscope and there is a nice saw but the frequence changes with the BC557b's temperature. I wonder if I can try to change the r12 value to put the transistor into a condition where it will drain less current and keeps it cold...

[PHOBoS]

if you encounter the same problem on all 4 builds (does the transistor also get warm on all 4 ?)
it seems less likely to be a component problem unless something was mislabeled and you used
the wrong one on all 4 builds. Did you use the Yusynth layout or build it on stripboard
or something else ?

[jansensouza]

I build it from yusynth layout
the point is for some reason the frequence is very dependent of the BC557b temperature and of course the current that flows trough it. I increased the R12 to something 39K, and the temperature seems to be ok and the frequence is about 100Hz maximum but it never stops to vary and when you change the Rate, the frequence does not stop, something like that: it is 100Hz now I change to 50Hz and then it begins to rise, if I bring it back to 100Hz it begins to fall, never saw something like that. It seems the transistor has a great sensibility for temperature, I know it does but seems to be excessive.

[PHOBoS]

ok well let's see what we can narrow down.

circuit design: I'd trust Yusynth designs to be working and several people have build it succesfully so can pretty much rule that out.

PCB design: as above since you used the same layout. if you etched it yourself there could be some errors but not very likely to
end up with the same error on all boards. Though if you use photosensitive PCBs and the same transparency it would be possible
that there is a small break in a track somewhere. In that case you could measure all the tracks to be sure they are conducting properly.

component failure: unless you got a bad batch of something this also seems unlikely. FETs can be suspect but 4 of them failing

wrong components: maybe somewhere something got mixed up and it's very easy to have the same mistake on all boards. Only way to
find out is to measure them all, at least the resistors. labels should suffice for the rest.*

wiring mistake: would also be easy to make the same mistake for all 4.

powersupply issue: You already measured that so unlikely. I don't know if you measured on the board itself after the 10 ohm resistors.
You might see something there but I really don't think that could be caused by the supply itself. Also the current does not seem to be
excessive.

magic: some behaviour can seem to be magical and unexplainable but in reality if it's not working properly there is a fault somewhere
and it should be possible to track it down. It can be rather difficult though and when you find it it may surprise you.


maybe there is something else I haven't thought about yet but I'd thoroughly check the components and check if you have the correct
values according to the schematic. Also check the wiring (again) against the schematic. Might be a good idea to draw it from what you
have and then compare that with the Yusynth schematic.




*I did see a post from Kevin Kissinger about some cap values here but I don't think that would cause the same problem.

[gabbagabi]

hm,
may it is something profan like bc557 soldered in the wrong way
or u have a bad/false bc557?

what else i would do in theory (of course)

solder out:
- the fet
- the range switch, put a jumper to on of the Caps, eg. J3a&J3b
do a test

solder out the the bc557
build the current scource on breadboard, and connect it
do testing

good luck

[jansensouza]

hi

I tried that way: first I rebild the circuit (only the oscilator section) on protoboard and the problem remains the same (almost like Led Zeppelin) . Then I realized that in fact the problem was that the current trough the BC557B was too high (for some reason) that lets it to get warm and as it gets warm the current (and the frequence) gets higher and the cycle begins again and again...

So what did I do:
1- put an 39K resistor on R12 instead of 27K to lower the current and do not change the Q1's temperature;
2- Change the capacitors C6a and C6b for the 10 times lower 0,1uF and 1uF (instead of 1uF and 10uF) to alow the transistor to charge them with 10x less current and
3- change the resistor R11 to 140K (100k+39K) to realign the slope of the frequence's curve and...

Voilá

Now I can put the frequence at 1x range from 100Hz to 0,1Hz and at 0,1x range to 10Hz to 0,01Hz with good precision and without drift. All four LFOs are now working equally well.
It costed me a good time to realize and I thank you all for all sugestions.
Now I'm gonna make the VCOs work properly (all 4) and I think it will be another adventure but now I have you to help me

Thanks in advance

[blue hell]

Great!