ICL7660SCPA troubles :(

[vigilante397]

So I've been working on an unnamed ( ) circuit that's been giving me a headache for a couple months now and I can't seem to get it right. I've tried three different build starting from scratch and nothing seems to be working. My most recent build I was actually able to locate a problem: my ICL7660 is getting REALLY hot. And I don't mean a little warm like some IC's do, I mean I still have a blister on my finger from where it burned me. At first I thought this was because I was using a 12V 1A power supply because the datasheet lists the maximum input voltage as 12V. So I tried it with a 9V 500mA power supply and it still burned me.

Is there anything else I could try to determine whether or not this is my problem?

[Mustachio]

Should take voltages on the IC and see if its putting out what you would expect. Also have you compared the power section to other circuits that use the same charge pump chip ?

Flanger ?

From what you've told us I don't think its the power supply but something in the power section of your circuit. Watch them fingers , its hard to rock'n'roll with burnt digits .

[italianguy63]

Yeah something for sure is wrong.  Make sure the output side of the chip isn't grounding... Definitely should not be hot.

[darron]

Did you replace the IC after you cooked it?

Those charge pumps are very, very sensitive to anything outside the ideal scenario. If you had a 12V 1A power supplyplugged in and you were only drawing a few milliamps then your supply voltage would certainly be greater than 12V. They only have a very tiny output current so if you accidentally short anything that shouldn't then it can be end of days too.


But yeah, we can't just take guesses all day. Get your metre out and see what the voltages are?

I accidentally put a max1044 on 11.3V for a moment not too long ago and POP. The 7660S seems a little better in this area though

[vigilante397]

I did replace the IC after I cooked it, and when I plugged it in again to check the voltages it started to heat up and smell like melting plastic (the first one actually melted my IC socket, so that was a mess to clean up) again so I unplugged it as quickly as possible. But I was able to get the voltages:

1: +9.225
2: +4.091
3: +0.007
4: -0.481
5: -1.160
6: -1.190
7: +5.190
8: +9.290

Thoughts? Does anything look amiss? Are there some things that look right and others that don't?

italianguy63: which pin is the output pin? That being grounded definitely sounds like something I don't want

[italianguy63]

Pin 5 is output.

https://www.intersil.com/content/dam/Intersil/documents/icl7/icl7660s.pdf

3 and 6 should probably be to ground.  The S model uses Boost, so 1 and 8 can be connected together... +9v

If you are doing basic voltage inverting.. Pin 5 should be around -9V.

MC

[armdnrdy]

First off...some information.

Are you using the 7660 to create a bipolar voltage?

If so...do you have access to a dual voltage power supply? +V/0/-V

If the answer is yes...you can remove the ICL7660SCPA from the socket (if the flames haven't already) and insert the +V wire from your supply to pin 8 of the socket. -V from supply goes to pin 5 and GND from supply goes to pin 3.

You can test the actual effect circuit by bypassing the charge pump.

Now I see that you are being secretive about what you are working on.....if you can't provide schematics, (or partials) and board shots, how do you expect someone to help with your problem?

I know of no clairvoyant members of this site. 

[vigilante397]

I guess I may as well come right out with it. I'm working on the Madbean Sunking, and I'm being secretive about it because I know the Klon (or Klone) can be a sensitive topic. According to forum rules I'm not allowed to post a schematic, but hopefully this information helps. 

[armdnrdy]

Did you etch you own board?

[vigilante397]

No, I purchased a pre-etched board from a forum member with way more etching experience than myself. If it had been a board I etched myself I would have assumed it was a problem from a  sloppy etching job, but this was a very clean etch.

[armdnrdy]

Well...

Since the build is verified, and the charge pump is being used to double the voltage, and seeing your 7660 voltages, the charge pump doesn't seem to be doing it's job!

This does not necessarily mean that the problem is centered around the charge pump.

You could have a solder bridge elsewhere along the power supply traces.

Look for a solder bridge. I know you probably have.....but keep looking!

I don't know how many posts that I've read where the poster claimed, "it's definatly not a solder bridge"...."I went over everything numerous times"

And you know what the problem turns out to be. 

[vigilante397]

The guy I bought the board from said to check diode and capacitor orientation, which I have done and found nothings wrong, so in-depth solder bridge checking was going to be my next step. I took a brief look and didn't find any, but I know that just because you don't see them doesn't mean they aren't there. I'll spend the night with my x-acto knife and let everyone know how it goes 

[vigilante397]

Well I definitely found my bridge. I was going through blindly with my thinnest x-acto knife carving up the tracks like I would on veroboard, making sure there weren't any "invisible" solder bridges that were killing my pedal. Then I remembered that Mark (italianguy) mentioned that the ouput pin, pin #5, should not connect to ground. So I checked that and found my solder bridge. Calling it a bridge doesn't even do it justice, it was a solder continental drift. It was HUGE I'm at work right now and I left my iron at home today, but I'll re-solder that when I get home tonight and see what happens.

[italianguy63]

Glad you found it!   

Maybe you can help me fix my Dyna Comp!

[vigilante397]

Maybe you can help me fix my Dyna Comp!

I get off work around 7, swing by and I'll take a look 

[vigilante397]

Looks like it's not time to congratulate me yet. 

I took care of my solder blob and actually found another very small solder bridge afterward. But the thing still refuses to work. I get a steady hum whether a guitar is plugged in or not, absolutely no sound gets through past the hum, the knobs do not affect the hum at all, the LED turns on when I hit the switch but the hum remains the same.

Is this pedal really as good as they say it is, or is it time for me to give up and accept the fact that I will never get it? 

[italianguy63]

One step at a time.  At least you have voltage now.. and at least one tone (huummmmm).  You'll get it!  MC

[the dead rat]

i realize this thread is old as shit. My apologies for that upfront, but its the exact situation i KEEP coming across with the EXACT same chip. and I'm breadboarding shit when it happens. i realize there's gotta be some shit i am doing wrong here, but there's been a few very basic (negative voltage converter from its datasheet) circuits and this mother@#$%er gets HOT like dude says. i smell it almost immediately, and i dont  necessarily have time to get readings because that shits getting hot enough im thinking its gonna melt the breadboard. but the set-up is exactly like the datasheet. and ive used the same IC (the 7660scpa) without any issues in lots of shit. I was using a 9v 200mA power supply this time. only thing i was supplying power to was a fairly typical dual op-amp (TL082) cv input/processing stage.

ill keep digging around on here and online but just curious if dude ever figured out. like, what the shit?

[vigilante397]

i realize this thread is old as shit. My apologies for that upfront, but its the exact situation i KEEP coming across with the EXACT same chip. and I'm breadboarding shit when it happens. i realize there's gotta be some shit i am doing wrong here, but there's been a few very basic (negative voltage converter from its datasheet) circuits and this mother@#$%er gets HOT like dude says. i smell it almost immediately, and i dont  necessarily have time to get readings because that shits getting hot enough im thinking its gonna melt the breadboard. but the set-up is exactly like the datasheet. and ive used the same IC (the 7660scpa) without any issues in lots of shit. I was using a 9v 200mA power supply this time. only thing i was supplying power to was a fairly typical dual op-amp (TL082) cv input/processing stage.

ill keep digging around on here and online but just curious if dude ever figured out. like, what the shit?

I'm the guy that was having the problem 8 years ago, and to be completely honest I have no idea It was a Klone on veroboard, and I vaguely remember working on it, but I don't remember if I ever got it figured out.