GSSL HELP THREAD!!!

[Gustav]

I still am unable to find information in these forums about how to check the voltages.  Does anybody have a link?

My former post.

Google "pinout 79L12", check DC voltage on output pin with reference to ground. Repeat for the rest of the regulators.

Gustav

check DC voltage = set your meter to read DC voltage.

Gustav

 

[up_right]

Update: GSSL now works - and sounds clean and sweet.. There was a bridge on the control panel. 

Special thanks to Gustav.

 

[up_right]

Now that I've had a moment to sit down, I wanted to share my experience about building the GSSL.

I ordered the full kit from Gustav:

https://pcbgrinder.com/index.php?route=product/product&path=60&product_id=50

Upon arrival the kit was incomplete - it was missing the SC switch, the Bypass switch, nuts and bolts for the  frame, the fuse and a poly cap for the SC switch. I immediately wrote Gustav and he promptly replied and sent the items free of cost. They arrived only a few days later.

I managed to build the kit in about 20 hours. There was quite a bit of confusion understanding the 'old' assembly PDF, but last week I realised that Gustav had uploaded a 'new' assembly PDF - and the new one is MUCH easier to understand, and includes MUCH clearer instructions.

Once I got the unit built it sounded distorted and none of the controls worked. After troubleshooting I realised that I had reversed the 78L12 and 79L12 regulator chips positions on the board!!! I swapped them back. Upon firing it up, the unit passed audio but still none of the controls on the control panel worked. I troubleshooted more and realised that there was a small bridge on the control panel between the middle solder points where the breakout cable comes out. I removed the solder bridge a voila! the GSSL came to life.

Overall this was a fairly painless operation.

If I were to improve anything about the kit, I would include in the assembly PDF exact instructions about how to check for voltages on the board. This is obviously a key part of checking the unit before putting in the chips and the lack of instructions led to a great amount of confusion. I searched this forum and online in general, but I could not find the correct info. I really am a beginner, and the learning curve for building this perhaps is a bit too steep for someone like me. Luckily I have a friend in electronics who helped me diagnose the problem area.

I would like to give thanks to Gustav for his continued help  and positive vibes and to Jakob for putting this awesome design out there for us!

Looking forward to the next build... 

 

[cray]

..add further led's in series - that way you use the same current (and drop less power in the series resistor),  we don't have too much reserve here..

Jakob E.

Tried with two leds in serie.
Yes  it works but , 78l12  now is working at higher temperature.
the unit works fine , but not happy with this : /

Is a good Idea to put a 30v lamp  directly  connected to the secondaries?
my gssl is a Rev#7
So not 7812 mounted  as the rev#12 .

Any suggestion is welcome.
Thanks in advance.

 

[SpeedGeezerButler]

So, I've got my GSSL rev 11 up and running. Debugged various solder bridges and a cold joint.

Everything seems to work fine except I do experience the regulator latch-up issue about every 5-6 power ups.

I checked and it's the +15 rail that goes away. Regulator in the circuit is an ST brand.

I've read through the thread for fixes and it's either change brands or try the diode/cap trick.

What is the current thinking on fixing the latch up? Swap brands, diodes, something else?

And if it's the +15 volt rail, I should be swapping out the 7815, correct?

I'm so close to getting this project put to bed...

Also, when the +15 goes, so does the +12.

 

[Gustav]

So, I've got my GSSL rev 11 up and running. Debugged various solder bridges and a cold joint.

Everything seems to work fine except I do experience the regulator latch-up issue about every 5-6 power ups.

I checked and it's the +15 rail that goes away. Regulator in the circuit is an ST brand.

I've read through the thread for fixes and it's either change brands or try the diode/cap trick.

What is the current thinking on fixing the latch up? Swap brands, diodes, something else?

And if it's the +15 volt rail, I should be swapping out the 7815, correct?

I'm so close to getting this project put to bed...

Also, when the +15 goes, so does the +12.

12V supply is an extension of the 15V supply, so its only natural they both go.

Seems like you've already found the possible solutions, none of them are "better" per se, but if you swap the regulator, you could have another lazy one, if you place the diodes, you should be set.

Gustav

 

[SpeedGeezerButler]

12V supply is an extension of the 15V supply, so its only natural they both go.

Yes, I see that on the schematic.

Seems like you've already found the possible solutions, none of them are "better" per se, but if you swap the regulator, you could have another lazy one, if you place the diodes, you should be set.

Gustav

So, even though the problem is only on the +15 side, should I put diodes on both + and - side?

Also, do I need any additional caps in parallel with the diode(s)?

Thanks for your input! Appreciate it. The compressor sounds awesome and just this one last fix before I can put the lid on and bolt it into the rack.

 

[tzman]

I would strongly advice to just swap the regulator for a better brand. See post no 6482 of this thread for a picture of the 'wrong' regulator and a working one. That will 100% fix your problem; just don't swap it for the exact same brand...

 

[SpeedGeezerButler]

I would strongly advice to just swap the regulator for a better brand. See post no 6482 of this thread for a picture of the 'wrong' regulator and a working one. That will 100% fix your problem; just don't swap it for the exact same brand...

Thanks tzman. Interestingly, the regulator on the right in the pic (the working one) is the one I have installed and not working.

I've ordered another of the same brand. I didn't see other choices at Mouser for L7815CV. 

 

[Harpo]

I've ordered another of the same brand. I didn't see other choices at Mouser for L7815CV.

...so don't search for L7815 if you don't want a STMicro part. Search for 7815 in order to get 55 results for MC7815, UA7815, LM7815, ..., pick an availiable/stocked, in single quantities, probably in TO220 housing, maybe from TexasInstruments, Fairchild, OnSemi or whatever else manufacturer part and be done with it. Using the additional anti-latchup diodes won't hurt and might help the STMicro part (in this specific circuit and pcb layout) as well. Your opinion on this subject might vary.

 

[SpeedGeezerButler]

...so don't search for L7815 if you don't want a STMicro part. Search for 7815 in order to get 55 results for MC7815, UA7815, LM7815, ..., pick an availiable/stocked, in single quantities, probably in TO220 housing, maybe from TexasInstruments, Fairchild, OnSemi or whatever else manufacturer part and be done with it. Using the additional anti-latchup diodes won't hurt and might help the STMicro part (in this specific circuit and pcb layout) as well. Your opinion on this subject might vary.

Thanks Harpo -- I didn't realize the leading character indicated the manufacturer - doh! I've got another ST on the way as well as the diodes.

Do I need two diodes - one for the +15 AND the -15 rail? Even though only the +15 has the issue? Just curious as I've ordered 10 of them to have on hand.

 

[Harpo]

...so don't search for L7815 if you don't want a STMicro part. Search for 7815 in order to get 55 results for MC7815, UA7815, LM7815, ..., pick an availiable/stocked, in single quantities, probably in TO220 housing, maybe from TexasInstruments, Fairchild, OnSemi or whatever else manufacturer part and be done with it. Using the additional anti-latchup diodes won't hurt and might help the STMicro part (in this specific circuit and pcb layout) as well. Your opinion on this subject might vary.

Thanks Harpo -- I didn't realize the leading character indicated the manufacturer - doh! I've got another ST on the way as well as the diodes.

Do I need two diodes - one for the +15 AND the -15 rail? Even though only the +15 has the issue? Just curious as I've ordered 10 of them to have on hand.

In parts datasheets of L7800/L7900 and L78S00 series, STMicro already advice to use these diodes (plural, so for both +/- rails) against potential latch-up problems  if the load is connected between +Vo and -Vo (such as opamps and VCAs as the only non-negliable load connected to the +/-15VDC rails in this GSSL circuit). Other manufacturers parts might be lesser sensitive in this respect, although the use of the anti-latch-up diodes will help these as well. I'd only use the diodes, but as always YMMV.

 

[miszt]

very noob question here, received my GSSL boards today, and noticed they have white pads, not come across those before, but looked around online and found references to "immersion tin", is that what this is? or do i need to do something special with these pads?

 

[Gustav]

very noob question here, received my GSSL boards today, and noticed they have white pads, not come across those before, but looked around online and found references to "immersion tin", is that what this is? or do i need to do something special with these pads?

https://en.wikipedia.org/wiki/IAg#Advantages_of_IAg

Gustav

 

[miszt]

I went ahead and started, everything was going great, until i checked this morning, and discovered lots of pads had lifted up overnight, room does get very very cold at night, but i guess i overheated the joints or something and it doesnt seem like they are still connected to the traces; i'm hoping that I can bridge them, and that although the others look fine, they haven't also lifted :/ going to try some continuity tests and cross my fingers!

 

[Gustav]

I went ahead and started, everything was going great, until i checked this morning, and discovered lots of pads had lifted up overnight, room does get very very cold at night, but i guess i overheated the joints or something and it doesnt seem like they are still connected to the traces; i'm hoping that I can bridge them, and that although the others look fine, they haven't also lifted :/ going to try some continuity tests and cross my fingers!

Could you post a picture of the solder side of the boards in question?

Pads lifting up over night is a new one to me...Lifting off pads at all usually requires quite a bit of abuse.

I set my iron to around 350 celcius when using lead free solder. Turn it down to 320/330 ish for some parts.

A lot of people find this helpful https://pcbgrinder.com/index.php?route=product/category&path=82_108

Gustav

 

[miszt]

I went ahead and started, everything was going great, until i checked this morning, and discovered lots of pads had lifted up overnight, room does get very very cold at night, but i guess i overheated the joints or something and it doesnt seem like they are still connected to the traces; i'm hoping that I can bridge them, and that although the others look fine, they haven't also lifted :/ going to try some continuity tests and cross my fingers!

Could you post a picture of the solder side of the boards in question?

Pads lifting up over night is a new one to me...Lifting off pads at all usually requires quite a bit of abuse.

I set my iron to around 350 celcius when using lead free solder. Turn it down to 320/330 ish for some parts.

A lot of people find this helpful https://pcbgrinder.com/index.php?route=product/category&path=82_108

Gustav

pics below (the 2 on the left circled by red dots have been cleaned up and bridged (the dirty brown mess on the pcb was from the "cleanup" and trying to bridge it) - the 2 on the right (just with single red dot) seem to be lifted aswell, but haven't got time to decide what to do re bridging those now)

i think the problem wasn't actually the soldering after all (apart from one joint they all flowed nicely within 1-2seconds), i think the problem was my cutters are blunt, and it yanked the component as i tried to cut it, just didn't notice until today when the same thing happened again, but this time without breaking it fortunetly

taken extra care finishing off the last few components today and all seems ok so far (except for the mess i made bridging the two broken ones lol eep ah well...its underneath, no one has to know)

pretty much finished now, although realized i have the wrong fuse, so testing will have to wait (typical lol)

 

[SpeedGeezerButler]

In parts datasheets of L7800/L7900 and L78S00 series, STMicro already advice to use these diodes (plural, so for both +/- rails) against potential latch-up problems  if the load is connected between +Vo and -Vo (such as opamps and VCAs as the only non-negliable load connected to the +/-15VDC rails in this GSSL circuit). Other manufacturers parts might be lesser sensitive in this respect, although the use of the anti-latch-up diodes will help these as well. I'd only use the diodes, but as always YMMV.

I used 2, 1n4001 diodes as described in the STMicro datasheet -- problem seems to be solved. I've power cycled about 20 times without a single latch-up. Before adding the diodes, I'd get a latch-up every 5 times or so.

http://datasheet.octopart.com/L7815CV-STMicroelectronics-datasheet-8222150.pdf (page 30)

Thanks to all for your guidance. Any plans to add these diodes to the next rev of the PCB? Thinking of future builders....

 

[senor_roboto]

GSSL current draw with DBX 202c gold can VCAs:

hey guys! long time lurker, first time poster here. I've built a couple GSSL's successfully so I'm familiar with the circuit and how the power supply works etc. does anybody know what the overall current draw from a GSSL with DBX 202c VCAs is?

I have a studer 369 console with two 1.915.100 power supplies and I was wondering if it's possible to install a GSSL off the power rails of the studer. there's a couple of empty spaces in the frame and i thought it would be cool to build in a better mix bus compressor.

the various rails coming off those supplies are
1.  +/-15v 1.5a max
2.  unregulated DC 30v 0.5a max
3. 48v phantom 200ma max

seems like the 30v dc unregulated could work right?

 

[miszt]

as the GSSL's onboard power supply provides +/-15V and +12V, wouldn't it make more sense to run it off the +/-15V rather than +30V unregulated?

that way you can you skip the AC conversion & 7915/7815 section on the board, and just have the 7812 to provide the +12V

seems simpler solution to me? no current worries then...unless DBX requires more than 1.5A (no idea, but if it does then obv the studer power supply isn't going to work anyway...)