FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 

SVF doesn't work (+15v / -15v are shorted or something)
MUFF WIGGLER Forum Index -> Oakley Sound Systems  
Author SVF doesn't work (+15v / -15v are shorted or something)

I just finished building a State Variable Filter module. So I plugged it in, but at that moment the other modules stopped working. So I rechecked my wiring, and everything seemed fine.. But still.

So I took out the SVF, and began measuring. Well, the +15V and the -15Vinput are connected with each other by accident. Zero Ohms resisance between those lines. I checked for bad solder joints, shorts, tin particles etc but found nothing. I removed the IC's, checked the placement of the caps, diodes and the 2 transistors , and it's still got this problem.. But everything looks to be okay.. confused

This will be hard to give some assistance, I think, but are there some likely faults or something that will have this shorting result?

I am presuming that there is no short between +15V and 0V and -15V and 0V?

If so there aren't many parts that straddle the power rails. Indeed, only the two offset trimmers, the tune trimmer and the frequency pot do so. A horrible failure in one of these parts would cause the problem you are seeing. Remove these parts one by one and see if the problem goes away. I'd check the trimmers first.

Another possible cause is a PCB short. This is where two tracks that aren't supposed to be joined are connected together somewhere. This is very unlikely since I use a very large gap between all my tracks and the boards are electrically tested. But, I guess it is possible.

So the places to look will be where the two rails run close to each other. Unfortunately, that will be the top side of the board and they run under the ICs. The short, if it is there, will be probably very thin. Now such faults normally blow out on the first application of power so if there is still one there it may be more obvious than just a fine hair's width of a track.

The point where the tracks are closest is under the OFF2 trimmer. The lower right pin of OFF2 just next to the legend U1 has a track very close to it. It may be worth looking at that gap once OFF2 has been removed.

Indeed, no short between +-15V and 0V.
I checked the offset trimmers and found out that Offset 1 was in its minimum resistance position, in other words fully counterclockwise. I set the trimmer to the middle position, and measured the +15V and -15V inputs. The short disappeared!

With the Offset 1 trimmer in the maximum position, the resistance between +15V and -15V is around 18kOhm with an unconnected SVF module. By adjusting Offset 1, the resistance decreases, ending with 0 Ohms (thus shorting) in the minimum position confused

Adjusting Offset 2 does not show this behaviour..
So I tried the module with the Offset 1 trimmer at the non-shorting position, and with the power on, the filter self-oscillates into a decreasing frequency. It does not pass external audio. But also, Q1 and Q2 are running hot, so I killed the power. Q1 and Q2 are the right types, as named in the builders guide.. Desoldered Tune trimmer, but didn't have any result. Frequency pot seems to be OK?
Check the right hand side solder pads belonging to R22 and R23. Make sure these are not bridged by a small sliver of solder either on the top or bottom side of the board.

Also a short between the right hand side solder pads of R22 and R21 will cook Q1 too. It will be worth measuring the resistance between these three pads as they should not be connected directly.

If no solder shorts could be found it could be a track short between those three pads too. This I would hope is a very rare occurrence since I have yet to hear of any trace shorts on any of my new boards from the current manufacturer. If it is a short then a small piece of copper will have been left on the board during manufacture. All the board shorts I have ever seen, bar one, have been very small. Some even burn out when you apply power. That this one hasn't indicates a solder short but it is possible to have a manufacturing defect that gives a more rugged short.

If it is a board short then simply take a sharp knife and drag it between the right hand solder pads of R21 & 22 and then R22 and 23. Do this on the bottom side and if that fails to remove the short do it again on the top side.

Hi, im back on track - went on holiday for a couple of weeks.

This SVF is still driving me crazy. There are no shorts between the right hand solder pads or R21/22/23. Also i can't find any track shorts.. I cleaned the underside of the pcb with alcohol to remove the flux (and mayble small solder slivers), but no luck at all.

I removed the frequency pot, nothing. Still shorting. With the Offset1 trimmer removed, the short is gone (47kOhm between +15 and -15 then).. But the trimmer itself is fine.

I also tried removing C9 and C11, but this didn't have any effect. The short must be somewhere further on the pcb, not at the power supply section..
It all sounds very frustrating.

With no power lead connected and the OFF1 trimmer removed - what is the resistance between the left hand solder pad (wiper) of OFF1 and...

1. 0V (pin 2 of PSU)
2. +15V (pin 1 of PSU)
3. -15V (pin 4 of PSU)

If these are all above 100K or so then the fault must lie with the trimmer. It is failing when the leads are compressed to fit in the board.

If any of the above register as below 150R then you have a short somewhere.

Thanks for the quick advice.

1.(0v) 490k
2.(+15v) 470k
3.(-15V) 500k

So it must be a faulty trimmer?

I've uploaded a picture. Without red drawn line on OFF1 there is no short between +15 and -15v. With these poles connected, the short is back. When I take look at the schematics I see that these poles are the +15v and -15v lines respectively. With a faulty trimmer the +15 and -15v will meet each other, and thus create a short?

Could I just connect a piece of wire between the wiper pad of OFF1 and one of the right hand solder pads?

With the wiper pad of OFF1 connected to the right hand upper pad, the unit passes audio. The LP/Notch/BP/HP outputs all sound as they should. The frequency pot doesn't affect the sound, only the resonance pot does so. And still the transistors are running hot..

With all the OFF1 solder pads unconnected, it behaves the same.

Edit 2: I measured the possibly faulty trimmer, but it seems okay? It goes from 0Ohm to 100kOhm, and the 2 'right hand leads' (with +15V and -15V) have a resistance of 100kOhm between them.
The offset trimmers are just there to reduce the unwanted DC offsets. The circuit will run without either of them, it will just have a small amount of DC voltage on its outputs.

The fact that the frequency pot does nothing and the two transistors are getting hot suggests that the two transistors are probably either swapped over or have failed.

Q1= BC560C
Q2= BC550C

I'm gonna desolder them one by one, and replace them with another transistor.

Audio output is also distorted on certain frequencies (when i sweep an oscillator), btw.
Yeah! We're coming closer. I replaced Q2 with another BC550C, and Q1 with a BC557. Now they aren't running hot anymore, and the Low Pass output works as it should. Frequency pot works, and the CV input too.

Let's see if it still works if i put back in that OFF1 trimmer..

Update: No more short with the trimmer put back in!

Is it ok to leave in the BC557 instead the 560? Or is the 557 inferior in this case?
Final post:

Weee! After tuning the filter every output works as it should, and now I have a great sounding filter.

Thanks for all the assistance in this frustating case.. (all caused by two bad very common transistors..)
MUFF WIGGLER Forum Index -> Oakley Sound Systems  
Page 1 of 1
Powered by phpBB © phpBB Group