MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

VC-LFO sync not working
MUFF WIGGLER Forum Index -> Oakley Sound Systems  
Author VC-LFO sync not working
renepela
I have an Oakley VC-LFO (Issue 1). Unfortunately the sync input is not working the way I think it should work.

Suppose I use a saw output on the LFO on a slow frequency.
I patch this output to the pith of my VCO.
Now I hear a tone go up slowly and at the ends it falls down and starts again.
When I also patch my gate signal to the sync input of the LFO I hope the saw wave starts at its beginning (lowest point) everytime I press a key on my keyboard.
Unfortunately it does not always happen (sometime it does). Some times When i press my keyboard the wave starts at its lowest point, but sometimes the saw keeps rising. The VCO does receive the gate signal, because I see the led flicker when I press the keyboard.

Can someone please give me some advice on how to troubleshoot this problem? What is the most likely spot on the pcb to check?

Any help is very much appreciated.

Rene
Synthbuilder
Yes, this is known problem with the old issue 1 design. There is some information about in the current issue 1 User Guide.

When I first designed the VC-LFO the DG403 I used was from Siliconix. However, Vishay bought out Siliconix and changed the spec of the DG403 - in particular the threshold on the digital inputs. This means that to provide guaranteed sync with the newer chips you have to provide 7.5V of gate signal and not just 5V. The older chips did work with 0V to +5V gate signals with no issue.

Solutions:

1. Amplify the gate signal with another module before sending to the VC-LFO. You do not need to amplify bipolar VCO outputs - these will sync just fine.

2. Find an older Siliconix and not Vishay DG403.

3. Mod your VC-LFO issue 1 accordingly:

Cut the track joining the +15V to pin 12 of U3. This is on the topside of the board and is under the IC. If you have socketed the chip, you may be still able to get to the track to cut it. I could on mine using the Rapid low profile DIL sockets, but you may have to take the IC socket off to do it.

Take two 10K resistors. For the first one, solder one end to pin 4 of U3 and the other to pin 12. For the other, solder one end to pin 11 of U3 and the other to pin 12. This will create a 7.5V supply for the DG403 on pin 12 which is its VL pin. This lowers the threshold voltage for the switch.

Make C2 a 150pF capacitor [5mm ceramic plate or dipped ceramic will do] instead of a 1nF. This maintains the pulse length which would be longer with the threshold voltage change.

I think Paul has been modding ready made VC-LFO modules with this mod for some time.

The problem has been corrected in issue 2 VC-LFO boards.

Tony
krisp14u
I have built about 10 with this mod and it is very simple to do and worked a treat

Mail me if you need some photos or help
numbertalk
Do you know when the issue 2 board was available for DIY? I ordered my VC-LFO PCB in September 09. I'm going to test this out tonight but without taking the module out of my cabinet (and being at work smile) wondering if I have an issue 1 board with this problem as well.
krisp14u
numbertalk wrote:
Do you know when the issue 2 board was available for DIY?

yesterday morning about 9:25 GMT Mr. Green
if you have the newer type DG403 you may want to do the mod
renepela
Thanks for all the help guys!
I'm going for option 3 and mod the lfo. Hopefully it will fix the problem.

Can you mail me a photo of the mod Paul?
numbertalk
renepela wrote:
Thanks for all the help guys!
I'm going for option 3 and mod the lfo. Hopefully it will fix the problem.

Can you mail me a photo of the mod Paul?


Or post here if you could - I'd like to see as well. Thanks!
krisp14u
I will post it here for all to see thumbs up

give me a little time as I will have to pull My VC-LFO out the rack
numbertalk
krisp14u wrote:
I will post it here for all to see thumbs up

give me a little time as I will have to pull My VC-LFO out the rack


Thanks Paul! Take your time.
numbertalk
Also, any advice on how to cut a trace? I've never actually done it - what's the cleanest/most effective way?
Synthbuilder
numbertalk wrote:
Also, any advice on how to cut a trace? I've never actually done it - what's the cleanest/most effective way?


I use a very sharp blade. One of them Xcelite types with a small wedge shaped blade will do. Score down on the track making sure you don't slip and go off and either cut another track or your finger. Keep on scoring over the same bit until the track is cut.

Now repeat the process on the same track but about 1 to 2mm to the side of where you have cut. This will create a dead area of copper that is now isolated from the two parts of the track.

With your soldering iron tip gently heat this island of copper track. You'll probably find you can actually scrape the copper away with the tip of your iron, but you can use the same knife blade to pick off the heated section of copper. This will now leave a nice 2mm gap in your track.

Clean the area with a little iso-propyl alcohol.

Alternatively you can use a disk sawblade on a Dremel. I don't like this method though - too much dust and not as accurate as a knife.

Tony
numbertalk
Thanks Tony! So it sounds like you also scratch off the solder mask for this section as well? I guess the copper layer is pretty thin, from what you describe?

Synthbuilder wrote:
numbertalk wrote:
Also, any advice on how to cut a trace? I've never actually done it - what's the cleanest/most effective way?


I use a very sharp blade. One of them Xcelite types with a small wedge shaped blade will do. Score down on the track making sure you don't slip and go off and either cut another track or your finger. Keep on scoring over the same bit until the track is cut.

Now repeat the process on the same track but about 1 to 2mm to the side of where you have cut. This will create a dead area of copper that is now isolated from the two parts of the track.

With your soldering iron tip gently heat this island of copper track. You'll probably find you can actually scrape the copper away with the tip of your iron, but you can use the same knife blade to pick off the heated section of copper. This will now leave a nice 2mm gap in your track.

Clean the area with a little iso-propyl alcohol.

Alternatively you can use a disk sawblade on a Dremel. I don't like this method though - too much dust and not as accurate as a knife.

Tony
Synthbuilder
numbertalk wrote:
So it sounds like you also scratch off the solder mask for this section as well? I guess the copper layer is pretty thin, from what you describe?


Just ignore the solder mask. The solder mask is not strong enough to cope with a direct assault from the soldering iron's tip or knife blade.

The copper is indeed very thin. But you need to heat it up to lift it. It's stuck down with a glue of some sort that weakens under heat.

Tony
numbertalk
Gotcha. So the mask won't harm my solder iron tip?

Synthbuilder wrote:
numbertalk wrote:
So it sounds like you also scratch off the solder mask for this section as well? I guess the copper layer is pretty thin, from what you describe?


Just ignore the solder mask. The solder mask is not strong enough to cope with a direct assault from the soldering iron's tip or knife blade.

The copper is indeed very thin. But you need to heat it up to lift it. It's stuck down with a glue of some sort that weakens under heat.

Tony
Synthbuilder
numbertalk wrote:
Gotcha. So the mask won't harm my solder iron tip?


I wouldn't have thought so.

But I abuse my soldering iron all the time. eek!

Tony hihi
numbertalk
Synthbuilder wrote:
numbertalk wrote:
Gotcha. So the mask won't harm my solder iron tip?


I wouldn't have thought so.

But I abuse my soldering iron all the time. eek!

Tony hihi


Thanks. Guess I'll find out how my iron does 8_)
renepela
Did you have time to make those photo's of the modification yet?
krisp14u
Sorry for the delay I can’t find my Photoshop disk and in the end found some software on the misuses laptop to do it

I'm nicking Tony's text


Synthbuilder wrote:

Cut the track joining the +15V to pin 12 of U3. This is on the topside of the board and is under the IC. If you have socketed the chip, you may be still able to get to the track to cut it. I could on mine using the Rapid low profile DIL sockets, but you may have to take the IC socket off to do it.




Synthbuilder wrote:

Take two 10K resistors. For the first one, solder one end to pin 4 of U3 and the other to pin 12. For the other, solder one end to pin 11 of U3 and the other to pin 12. This will create a 7.5V supply for the DG403 on pin 12 which is its VL pin. This lowers the threshold voltage for the switch.






Synthbuilder wrote:

Make C2 a 150pF capacitor [5mm ceramic plate or dipped ceramic will do] instead of a 1nF. This maintains the pulse length which would be longer with the threshold voltage change.


hope this makes it a little clearer

thanks to Mr B Dunnett for the photo of the top side
numbertalk
Thanks Paul!

Question - this mod won't make this chip more sensitive to higher voltages plugged into the sync jack, will it? My CV keyboard has 10VDC gate and trigger signals - these should be fine even after the mod, right? (Why am I modding mine then, you ask? Because I have other sequencers and whatnot that don't put out quite as hot of a gate/trigger).
Synthbuilder
numbertalk wrote:
this mod won't make this chip more sensitive to higher voltages plugged into the sync jack, will it?


Sensitive as in harmful? Probably not.

The specs on the DG403 are not very clear in this respect. The data sheet says that the max signal input should not exceed +/-14.4V. And the internal schematic seems to suggest this to be the case. However, it doesn't say what effect voltages above VL - the supply voltage to the DG403's logic devices - will have on the actual device's switching behaviour. Probably nothing...

But if you do have some odd behaviour when using high level triggers I would add this to the modifications given above:

Make D1 a 6V2 zener diode instead of the BAT42.

Tony
numbertalk
Thanks - good to know. I have the means to easily amplify signals that are too low to trigger sync, so maybe I'll just do that.

Synthbuilder wrote:
numbertalk wrote:
this mod won't make this chip more sensitive to higher voltages plugged into the sync jack, will it?


Sensitive as in harmful? Probably not.

The specs on the DG403 are not very clear in this respect. The data sheet says that the max signal input should not exceed +/-14.4V. And the internal schematic seems to suggest this to be the case. However, it doesn't say what effect voltages above VL - the supply voltage to the DG403's logic devices - will have on the actual device's switching behaviour. Probably nothing...

But if you do have some odd behaviour when using high level triggers I would add this to the modifications given above:

Make D1 a 6V2 zener diode instead of the BAT42.

Tony
renepela
Thanks for the help. I just made the change to my lfo and it's a bit better now.
Still in some cases I can't make the vco wave start at's beginning, but I guess the sync just doesn't work as I think.
Synthbuilder
renepela wrote:
Still in some cases I can't make the vco wave start at's beginning, but I guess the sync just doesn't work as I think.


Have a look at the issue 2 User Manual. I have added a much more detailed section on how sync works for the various waves. These details are mostly the same as the modified issue 1 VC-LFO and should help you understand the operation of the rather complex sync process. You should note though, that both the triangle and sine waves will rise after a sync pulse in the issue 1 VC-LFO while they fall in the issue 2.

Tony
MUFF WIGGLER Forum Index -> Oakley Sound Systems  
Page 1 of 1
Powered by phpBB © phpBB Group