jooks Messages: 21 Registered: December 2009
Location: sweden
Junior Member
just finished building the bend matrix and hooked it up for voltage check. No IC´s at all. I have used 330 ohm resistors as per BOM instead of 220 ohm as per pcb legend. The 100pf cap is a mica since that was all I had. Powered with unregulated 9vdc, reads 12.18V going in. I get the following voltages:
ATM32
PIN 10: 4.99V
PIN 11: 0.00V
PIN 30: 8.15V
PIN 31: 0.01V
MT8816
PIN 40: 4.99V
PIN 20: -0.08V
PIN 16: 0.00V
I get 4.99V at all the extra +5v places on the pcb. I also get 0.00V but at -5V I get -0.08. So ATM pin 30 gets 3V too much and I get -0.08V instead of -5V....To me it seems that this don´t have anything to do with using 330ohms instead of 220? I used headers and a shorting bar for the 5v/9V option so it could easily be changed. Maybe it´s not contacting as good as it should but i´ve never had troubles with that before. Not sure were to start looking here...
Any help would be much appreciated! Thanks,
Erik
jooks Messages: 21 Registered: December 2009
Location: sweden
Junior Member
Ok, sorry for writing before thinking. I had forgotten to complete the circuit for negativ voltage (IC 7660 and 4049) so that explains the lack of -5V.
Voltage check confirms -4.68V so not quite -5V. Is -4.68V close enough to work?
I also still get 8.15V for ATM32 PIN 30. What´s up with that?? +4.99V and 0.0 at all the other places...hmmm
jooks Messages: 21 Registered: December 2009
Location: sweden
Junior Member
Ok, fr.o.m. Datasheet of atmega32 Pin30 is AVcc. From BM schematic Avcc seems to be getting 5v from the second 7805 (to92). F this is correct then the problem might be with that regulator or something nearby?
Edit:
I get 8.91V into the 7805 and 8.14V out from it (exactly the same as on pin 30). Next step order a new 78L05 and replace..
jooks Messages: 21 Registered: December 2009
Location: sweden
Junior Member
looking at The datasheet for the 7805 (national) pinout might be reversed compared to what to92 regulators usually have. Input is on the left standing frnt of the flat side. I believe this is reversed from the bend matrix? I will check this out as soon as I can.. . Seems plausible from the voltage readings anyway.
jooks Messages: 21 Registered: December 2009
Location: sweden
Junior Member
The pinout was indeed different on the 7805 regulator (to92) so I resoldered it the opposite way. It now puts out 3.93V....it gets 8.92V in from the 7809. Is it possible that it was damaged from the desoldering? I cut the leads and pretty much surface mounted it just to see if it would work (I have a new regulator coming in the post).
jooks Messages: 21 Registered: December 2009
Location: sweden
Junior Member
Yeah it´s probably broken but the strange thing is I got 2 new ones and I have now tried both of them flipped the "right" way but again I get 8.92V in and 8.14V out. At least it´s consistent... maybe I got a different brand this time (from same supplier) and again with a different pinout then I thought??
Any ideas what it could be if not the voltage reg?
Hmm, perhaps a random blob of solder bridging a connection somewhere?
Make sure you're actually measuring your voltage between gnd and the "out" pin.
As far as I'm aware, all of the TO-220 7805's should have the same pinout. The difference is between the TO-92's and the TO-220's.
You could also try removing U700/U701 and the MCU if they're installed as they seem to be the only thing I can see on the schematic that AVcc feeds... I can't envision an opamp being capable of this, but if there's a bridged connection somewhere, I guess it's possible. And check continuity between the different vregs and make sure nothing's connected that shouldn't be. Without looking at the datasheet of the MCU I believe the Vref pin shouldn't like seeing anything above 5.5V or so. I'm hoping you don't already have those chips plugged in - if so your knob effects may not work or the whole chip may be fried.
Try feeding it +9V/GND out of the circuit and measure.
jooks Messages: 21 Registered: December 2009
Location: sweden
Junior Member
I haven´t discovered any shorts so far (but not 100% ruled out). I have checked the Avcc voltage both on the regulator output and MCU so it should be ok (well, not ok)
Yeah the to220 and to92 are different but from what I could google the to92 I used had a different pinout then what is specified on the bend matrix pcb legend. I will get some more and try again....
Luckily I have no IC´s mounted except for the neg.voltage supply.
many thanks for the help! I´ll get back with a report when I have tested some more.
jooks Messages: 21 Registered: December 2009
Location: sweden
Junior Member
Still no luch with this. I have 8.91V into the AVcc regulator and 8.19V out....I´ve tried 4 different regulators. Pinout is right. No shorts from what I can measure. I really have no clue to what the problem is??? The regulator gets 8.91V in which should be ok and it gets ground and no shorts between the pins so what can be wrong? The regulator only takes 0.1mA but that should not be a problem I think? It´s a national LM78L05. Should work or?
Any help much appreciated!
LM78L05 should work just fine. With it removed and power applied, check to see if the positive side of the cap or the hole for pin 3 of the 7805 has any voltage there... That could indicate accidental coupling to some other section of the circuit.
jooks Messages: 21 Registered: December 2009
Location: sweden
Junior Member
I´ve checked for any voltage on pin 3 of the regulator and there is none....I´m about to give up. The regulator gets 8.9V, have connection to ground on middle pin and no shorts and still 8.2V out with every regulator I´ve tried (and I´ve tried quite a few now)
dann Messages: 829 Registered: December 2002
Location: Austin TX
Senior Member
Hey, try simplifying your system: take the bend matrix out of the picture and see if you can just get a 7805 to produce 5V. Perhaps use a breadboard and run 9V into a 7805, and measure the output. If you can verify that your 7805 will regulate voltage properly then put it the bend matrix and see if it's still working