| Author |
Message |
Lemmy
Joined: Apr 12, 2007 Posts: 106 Location: London
Audio files: 1
|
Posted: Mon Jul 09, 2007 12:36 pm Post subject:
|
 |
|
| Randaleem wrote: | Dang! this one's buggin' me!
I KNOW it's gonna be something easy; but I just am not seeing it.
BTW, your build looks nice! As a self-proclaimed newbie; You've done a great job.
|
Thanks! I need some encouragement at this point
| Randaleem wrote: |
1) Check to be sure the heatsinks are not bridging the pins of the 79xx vregs (Check both. Use your meter to each pin on the trace side; measuring to the heatsink.
2) Lay it on it's side and measure the voltages; so whatever it's been sitting on is not causing something to act funny or short out. (Clip those long leads.)
3) The one that's been bothering me the most for the longest. Are you using a 9V transformer? Because the 30V you've mentioned from Vreg inputs to GND seems excessive. And that could be the problem. Because you might be over-stressing the 7909 and/or the 7912?
But I have to say that this is a weird one. I get the funny feeling I'm goinna be embarrassed when this gets sorted out somehow
|
It is a 9v transformer.
I'm sure you're right, it's probably something dead simple.
When I touch my black probe to ground, and the red probe to the 7912 heatsink, I get a -30v reading. Is that right? Should the back of the VR conduct the input voltage? I get the same when I put the red probe to the 7909 heatsink. When I put the probe to the other two heatsinks, I get a 0v reading. Confusing.
Anyway, I removed the heatsinks from the 7912 and 7909 completely, and that didn't make any difference.
One other thing - I sprayed the copper side of the board with some solder flux spray after etching it. AFAIK this is to stop the the copper oxidising, and it is supposed to help when soldering in components. I found it to be funny stuff though... it sealed the board but made soldering more difficult - the copper on the board wouldn't heat up so well. Any chance that I did something wrong in using that stuff? Could it have made some bridges somewhere?
Would having a high negative voltage cause potential problems, by the way?
Thanks. |
|
|
Back to top
|
|
 |
Randaleem
Joined: May 17, 2007 Posts: 456 Location: Northern CA, USA
|
Posted: Mon Jul 09, 2007 4:18 pm Post subject:
|
 |
|
Lemmy,
Last things first. YES, having the -12V of your synth actually be -21V is a BAD thing. Many circuits designed for +/-12 or +/-15V would be destroyed if powered by +12/-21 !
Many IC's have a limit for the total voltage across their plus and Minus supply pins. You'd be giving them 33V, and they are expecting 24 or 30. Many common CMOS chips in synths have a limit of 18V from supply to ground. Being creative, synth circuit designers will often use chips in creative ways. Sometimes this means that a chip gets used across one of the supplies. If one of these 18v chips was placed across your -21V , it would smoke instantly!
Worse, the resistors values in a circuit are calculated based upon an assumed supply voltage. So if you put 21V into a circuit designed for 12 or 15 the resulting voltages and currents thoughout the circuit will not be right. Now in some cases; with some circuits; this wouldn't matter. But you're dealing with a lot of variables here. Let's get this power supply right. It's important to your further success!
Okay, to your questions...
Here's a link to the datasheet for the 7912:
http://www.national.com/pf/LM/LM7912.html
If you look to the right at the drawing of the package you'll see that the mouting tab is indeed at the input voltage. So that's okay. And here:
http://www.fairchildsemi.com/ds/LM/LM7812.pdf
we see in figure 2 that the positive 78xx series does have its mounting tab at GND. So that's okay as well.
BTW, getting into the habit of looking up the datasheets for parts in your circuits is IMO a good idea. Having a datasheet on hand for the parts you use will be one of the best investments in this hobby that you can make.
But I'm still concerned about why you're getting 30V instead of something like 25-26? This is a voltage doubler circuit, and you're driving it with a 9VAC transformer. So 9 x1.414 (to get the peak value of an AC voltage measured with a DVM/MM/VOM you multiply it by the square root of two, which is 1.414.) As I wrote above somewhere, the measured AC is about 70% of its peak (actually 70.71%, which is the inverse of 1.414)
Point here is when we double that (voltage doubler circuit) we should be getting 25.45V or thereabouts. And you're measuring 30. (Now the 7912--from the datasheet!<G> says we're okay to use down to -35V; so this isn't our problem. But when things aren't working right, IMO it's a good idea to look at EVERYTHING that seems even a "little" off.
Having said that, if your "9VAC" transformer is designed for a 110V primary (older transformer) and you're feeding it with 120 or even 125, that answers the question. Because then you'd have 125 over 110, or a 1.135 multiplying factor. 9 times 1.136= 10.227VAC actual. Times 1.414=14.46. Times doubler circuit= 28.9. Which is a lot closer to the 30V you measured.
If you happened to buy a 105VAC primary transformer (which is not too uncommon from some electronic surplus places) then you're at 30.3V worst case with 125 mains. (While I'm using USA mains volts examples the Mains voltages in other countries has similarly risen over the years; 215 to 240 for example. So the effect of using older transformers or gear at newer voltages is something to keep in mind with synths; both DIY and otherwise. )
I Hope you don't mind getting the explanation and not just the "answer". The idea is to "talk it out" and show he steps of the process; so people reading this might be able to solve similar but different problems .
Okay, NOW we might be getting somewhere. You mentioned spraying a "flux" and "copper protector". This is a very useful piece of information. Do you still have it? I'd like to know what it's called?
Usually you use a flux to solder and then eitehr wash it off or its a type that doesn't need to be washed off. And if you're wanting to protect the copper, you either spray it after assembly with a conformal coating (some poeple use Krylon clear spray, and swear by it) or you cold plate it with Tin before soldering (Tinnit and Liquid Tin are two trade names.).
A combination flux/ conformal coating seems an odd duck. But we live in amazing times; so maybe its really cool stuff?
Here's how it may have affected your situation; and this is all guessing since I don't know the product specifics.
A) Regular flux should be cleaned after you solder because it can act like a resistor between traces that are supposed to be isolated from each other. So if your spray is doing the same thing; it might be a clue.
B) A conformal coating can sometimes get into the connection between the component lead and PCB trace; making an open or increased resistance pathway. (You run into this sometimes with rework of older conformal coated boards.
I know if I were you right now I'd be looking into exactly what the stuff is and how it might be involved.
We have another avenue that might be easier than trying to remove all the special coating (Since I don't know what it is; even suggesting its removal would be out of line, IMO.) Here's the other avenue:
I mentioned earlier that you might want to remove the 7909, since you've said that your original reason for adding it was the thought that the SL and the add-on would need both voltages. If you remove it now., we'll know that it cannot be contributing to any offset in the 7912 output.
That's what I'd do next...
I'd remove the 7909, and re-measure the remaining three voltages. If the -12 is still at -21V ; I'd go ahead and remove the 7809 vreg too. But do the 7909 by itself first. Taking two out at once won't tell us what happened if the problem goes away.
If it's STILL -21V with both 9v Vregs gone, then it HAS to be something else.
Randal
| lemmy wrote: | | Randaleem wrote: | Dang! this one's buggin' me!
I KNOW it's gonna be something easy; but I just am not seeing it.
BTW, your build looks nice! As a self-proclaimed newbie; You've done a great job.
|
Thanks! I need some encouragement at this point
| Randaleem wrote: |
1) Check to be sure the heatsinks are not bridging the pins of the 79xx vregs (Check both. Use your meter to each pin on the trace side; measuring to the heatsink.
2) Lay it on it's side and measure the voltages; so whatever it's been sitting on is not causing something to act funny or short out. (Clip those long leads.)
3) The one that's been bothering me the most for the longest. Are you using a 9V transformer? Because the 30V you've mentioned from Vreg inputs to GND seems excessive. And that could be the problem. Because you might be over-stressing the 7909 and/or the 7912?
But I have to say that this is a weird one. I get the funny feeling I'm goinna be embarrassed when this gets sorted out somehow
|
It is a 9v transformer.
I'm sure you're right, it's probably something dead simple.
When I touch my black probe to ground, and the red probe to the 7912 heatsink, I get a -30v reading. Is that right? Should the back of the VR conduct the input voltage? I get the same when I put the red probe to the 7909 heatsink. When I put the probe to the other two heatsinks, I get a 0v reading. Confusing.
Anyway, I removed the heatsinks from the 7912 and 7909 completely, and that didn't make any difference.
One other thing - I sprayed the copper side of the board with some solder flux spray after etching it. AFAIK this is to stop the the copper oxidising, and it is supposed to help when soldering in components. I found it to be funny stuff though... it sealed the board but made soldering more difficult - the copper on the board wouldn't heat up so well. Any chance that I did something wrong in using that stuff? Could it have made some bridges somewhere?
Would having a high negative voltage cause potential problems, by the way?
Thanks. |
|
|
|
Back to top
|
|
 |
Lemmy
Joined: Apr 12, 2007 Posts: 106 Location: London
Audio files: 1
|
Posted: Wed Jul 11, 2007 9:24 am Post subject:
|
 |
|
| Randaleem wrote: | Lemmy,
I Hope you don't mind getting the explanation and not just the "answer". The idea is to "talk it out" and show he steps of the process; so people reading this might be able to solve similar but different problems .
|
No problem, lots of useful info here, appreciate it!
Well, we're narrowing it down. I took out the 7909. Still getting -21v out of the 7912. Took out the 7809. STILL getting -21v out of the 7912.
Onto the flux spray:
| Randaleem wrote: |
Okay, NOW we might be getting somewhere. You mentioned spraying a "flux" and "copper protector". This is a very useful piece of information. Do you still have it? I'd like to know what it's called?
|
The stuff is here, described as PCB Flux Spray: http://www.users.zetnet.co.uk/esr/PCB_Chemicals.PDF
On second look it seems to be appropriate for how i used it.
My 9v power adapter is brand new by the way, and I'm running it in the UK, so at 240v.
Maybe I just need to get hold of a new batch of 7912s  |
|
|
Back to top
|
|
 |
Lemmy
Joined: Apr 12, 2007 Posts: 106 Location: London
Audio files: 1
|
|
|
Back to top
|
|
 |
Lemmy
Joined: Apr 12, 2007 Posts: 106 Location: London
Audio files: 1
|
Posted: Sat Jul 14, 2007 6:55 am Post subject:
|
 |
|
Elsewhere on the internet, someone told me that the 7912 needed a minimum load, and then it would work OK. As a newb I'm not sure how to act on that advice...
It seems like the other posts I'm reading where people are having similar problems involve 7912s manufactured by ST, which is the kind I have. So I've got some different ones on order. I hope that might sort the problem out. |
|
|
Back to top
|
|
 |
Lemmy
Joined: Apr 12, 2007 Posts: 106 Location: London
Audio files: 1
|
Posted: Fri Jul 20, 2007 3:41 am Post subject:
|
 |
|
It worked with the new 7912...!
Phew.
Thanks for all the help everyone. |
|
|
Back to top
|
|
 |
Randaleem
Joined: May 17, 2007 Posts: 456 Location: Northern CA, USA
|
Posted: Fri Jul 20, 2007 2:08 pm Post subject:
Subject description: Hooray! |
 |
|
| lemmy wrote: | It worked with the new 7912...!
Phew.
Thanks for all the help everyone. |
Now THAT is some GOOD news!
   
Congratulations!
Randal |
|
|
Back to top
|
|
 |
Lemmy
Joined: Apr 12, 2007 Posts: 106 Location: London
Audio files: 1
|
Posted: Sun Jul 22, 2007 2:59 am Post subject:
|
 |
|
Ha ha, thanks!
And now on with the building of the mini-synth! |
|
|
Back to top
|
|
 |
Pehr

Joined: Aug 14, 2005 Posts: 1307 Location: Björkvik, Sweden
Audio files: 2
|
|
|
Back to top
|
|
 |
tommi

Joined: Dec 05, 2007 Posts: 247 Location: Italy
Audio files: 3
|
|
|
Back to top
|
|
 |
Danno Gee Ray
Joined: Sep 25, 2005 Posts: 1351 Location: Telford, PA USA
|
Posted: Sun Dec 23, 2007 8:01 am Post subject:
|
 |
|
| I have, and it does. 12V works just fine. |
|
|
Back to top
|
|
 |
tommi

Joined: Dec 05, 2007 Posts: 247 Location: Italy
Audio files: 3
|
Posted: Sun Dec 23, 2007 8:15 am Post subject:
|
 |
|
Many thanks Danno Gee Ray!
So, now I can relax and wait for PCB and components to come.  |
|
|
Back to top
|
|
 |
PoindexterLabs

Joined: Jun 07, 2007 Posts: 12 Location: North Carolina
|
Posted: Mon Mar 10, 2008 6:11 pm Post subject:
Amps and Barrel Jacks |
 |
|
Maybe someone addressed this and I just didn't see it, but is the amperage of the voltage regulators or the wall wort an issue? My VR's are 12V @ 1A. I have an inky feeling I should lower it to 500mA, but then again I'm just cresting this point between beginner and intermediate (i.e. enough knowledge to know when I'm potentially screwing up). Part of me feels like 1A is just too much current to be pushing in.
Also, has anyone used a barrel jack like the kind used on effects pedals instead of running directly off the wall wort wires (try saying that three times fast) to the Power Supply? (And by the by, I am using Ray's PSU.) I have more experience with building guitar effects pedals, so this is more familiar, and I would think it would make things a lot easier and more portable. But does anyone forsee a potential problem with using these jacks?
Anyone more knowledgable than I, please assist. |
|
|
Back to top
|
|
 |
PoindexterLabs

Joined: Jun 07, 2007 Posts: 12 Location: North Carolina
|
|
|
Back to top
|
|
 |
blue hell
Site Admin

Joined: Apr 03, 2004 Posts: 24493 Location: The Netherlands, Enschede
Audio files: 298
G2 patch files: 320
|
Posted: Mon Mar 10, 2008 7:34 pm Post subject:
|
 |
|
It does not matter how much current your power supply is able to deliver, as long as it's more than you need, or equal. The amount of current actually drawn will be determined by the voltage of the supply and the resistance of the consumer, as long as the supply is able to deliver that.
So when the resistance of the consumer is such that at a certain supply voltage 100 mA is drawn, the power supply must be able to deliver that. When it can do more the excess capability will not be used. It will not have to be dumped somewhere, it just doesn't flow.
Only thing is that when you would use a 10A power supply any accidental short circuit would be nasty, and it would be way to expensive as well for what it has to do.
Over voltage of the power supply is no good though, as that will be seen by the consumer and it might destroy some components not rated for the higher voltage.
V=I*R or I=V/R
V is the voltage of your supply, R is the resistance of the consumer. I, the current flowing through the consumer, is determined by V and R. The power supply has to be able to meet that demand. When it can do more that's fine. When it can do less the current can not flow (as it can't be delivered) and as a consequence V will go down (as R is a given), a short circuit situation is beginning to emerge, and both producer and consumer will not be happy. _________________ Jan
also .. could someone please turn down the thermostat a bit.
 |
|
|
Back to top
|
|
 |
PoindexterLabs

Joined: Jun 07, 2007 Posts: 12 Location: North Carolina
|
Posted: Mon Mar 10, 2008 7:48 pm Post subject:
|
 |
|
Thanks Blue Hell, but I'm only using a 1A power supply and not a 10A. And my wall wort is also only 12V so I think all specs are within parameters. I do appreciate your thorough explanation.
Do you think it would be a good idea to throw a fuse inline for good measure, or would that be overkill? |
|
|
Back to top
|
|
 |
blue hell
Site Admin

Joined: Apr 03, 2004 Posts: 24493 Location: The Netherlands, Enschede
Audio files: 298
G2 patch files: 320
|
Posted: Tue Mar 11, 2008 2:40 pm Post subject:
|
 |
|
A secondary fuse you mean?
I never used one, and probably the wall wart will be short circuit protected - at least for a short while. You could do it when you think the circuit connected to it might blow up due to a mistake. Others have used small resistors (like 22 E) in the power line(s) for that purpose. _________________ Jan
also .. could someone please turn down the thermostat a bit.
 |
|
|
Back to top
|
|
 |
A30N
Joined: May 06, 2009 Posts: 44 Location: Southampton, England
|
Posted: Fri May 08, 2009 4:30 pm Post subject:
|
 |
|
Hi,
Im a little stuck on sourcing The LM 7909, i have the other 3 regulators found but not this one.
Can anybody give me an alternative?
Thanks,
Jamie
P.s im not really fussed about the 9v can i just disconnect c 9-12/ lm 7909/7809 |
|
|
Back to top
|
|
 |
LetterBeacon
Joined: Mar 18, 2008 Posts: 454 Location: London, UK
|
Posted: Sat May 09, 2009 9:10 am Post subject:
|
 |
|
| Maplin sell the the TS7909 which I'm pretty sure works. |
|
|
Back to top
|
|
 |
A30N
Joined: May 06, 2009 Posts: 44 Location: Southampton, England
|
Posted: Sat May 09, 2009 12:31 pm Post subject:
|
 |
|
| LetterBeacon wrote: | | Maplin sell the the TS7909 which I'm pretty sure works. |
Hey thanks,
I think im going to go with another design in which i only need the 12v regs  |
|
|
Back to top
|
|
 |
LetterBeacon
Joined: Mar 18, 2008 Posts: 454 Location: London, UK
|
Posted: Sat May 09, 2009 1:29 pm Post subject:
|
 |
|
| No worries. Maplin also do the 12v regulators if you're looking for them too. |
|
|
Back to top
|
|
 |
|