The OVP is most likely implemented somewhere around the PWM, on the primary side - logic would dictate that the LM339 is most likely used for that (as well as other functions), so i'm afraid i'm / we're a bit stuck, regarding that...
Khron thanks for definitions.
And no my original idea was not to alter hot side.. I made a mistake, i assumed that preregulator was pwm controller after a quick look in the datasheet. Therefore i decided to hack its feedback, while my goal was to alter feedback circuit for pwm driver, exactly as it was done here.
Regarding OVP, LM339 does indeed have 4 comparators inside that are most likely for protections. From my experience in ATX modding i would say that altering feedback doesn't trigger OVP for some reason (i guess it changes the reference voltage which maybe comes from the feedback signal?). The zener diodes (or other crowbar protection components) can do it though, so check the output side as well.
Of course remember to check cap voltage rating on the output, as you don’t want to exceed those.
If there are no crowbar protections on the output, lm339 is really simple chip, what i would do is i would get some help (few more hands or so), and find out which one of 4 comparators are used for OVP by checking voltages on input pairs while increasing output voltage with trim pot.
find out which IN+ (or IN- but i don't think so) increases together with voltage, and OVP should trigger as soon as in+ is as high as IN- (for the same comparator).
The input pairs on LM are following: pins 4 and 5; 6 and 7; 8 and 9; 10 and 11; giving outputs to pins 2; 1; 14; 13, respectively.
pins 3 and 12 are Vcc and GND, respectively.
Then fake that voltage or disconnect input pin or something.
Thats just an idea. you don't want to disable OCP though
Best thing to do is to make psu shutdown at 15.5v for safety of capacitors and load, best voltage for powering car amp is around 15V. My idea is to increase resistance little bit at comparator 's voltage input, if it's posible.
I did not see what is the current OVP level? And cap ratings, are they 16V?
I don't see any reason for running car amp at more than 13-14V. For lead-acid battery it might be needed as voltage drops quite a bit when you load it. While this is a regulated p/s it should keep up with voltage at different loads unless you push it to the max.
i run my adjustable atx until 15,5V as well. I did not need to adjust anything on the comparator as it worked fine from the very start. I have limited it with a adjustable resistor so on extreme setting on trim pot (max voltage) it is 15.5V
I have measured voltages on LM339 and there was change only in pin 6, after disconnecting pin6 from board, no changes were made, I also tried to remove CXA8038's OVP control pin and same thing...ideas?
then please go back to testing LM339,
Output pins: 2; 1; 14; 13, please check their voltage in relation to GND.
Try checking one that changes when ovp is triggered.
then please go back to testing LM339,
Output pins: 2; 1; 14; 13, please check their voltage in relation to GND.
Try checking one that changes when ovp is triggered.
Already did everything, I will repeat just in case.
EDIT:
pin 1-0.14v
pin 2-0.06v
the following pins have a little bit change when I change resistance from 12v to 13.7v
pin13 0.04v and 0.05
pin14 3.56v and 3.58v
tracing, not measuring but if you want to measure them, check if those pin voltages get close to each other when getting closer to ovp level. Else try making very clear pictures of both sides of the board so some of us can try to look into that
Comment