Hello! I'm working on a project where I need to accurately switch between a -75v and a 75V rail. I've got the step-up circuit working fine, but I'm stumbling on controlling it. I was originally using two pins to switch some mosfets allowing me to pass +5 and -5 to the "pwm in" net-tag, but it doesn't really work well. Can anyone point me in the right direction for that?
The other thing I would like to do is either monitor the current going out from the "to output" net-tag (being able to control the current output would be nice, but I'm fine with just using resistors to do so), or add some form of continuity test to know when the output probe is connected.
I appreciate any help on this. I inherited this project recently, and it's based on a design from 1985 that where I work makes, and the original engineer is no longer around to ask questions to. Thus I'm trying to remake it in a more modern form, with parts that are currently available.
What is the intended PWM frequency?
Implementing current current control in that circuit might no be easy. It easily ends up in an oscillating output. A resistor is handy. Another current limiting device could be a lightbulb! They have a low resistance a low current but the resistance increases with an increasing current. A 110 volt bulb of a selected Watt rating is thinkable.
One difficulty is that the PWM needs to swing between -5 and +5 volt.
Show us. What was wrong with it? What was the source of the PWM signal? You didn't say anything about that. Do you need to keep or replace that source?
Calling it PWM is honestly kind of inaccurate on my part, it's more just switching between the two supply rails at .3, .6, 7.83, 20, 40, 200hz via turning one output pin on to switch on the supply rail for that spot. My circuit worked in sim, but in hardware it just overheats quickly (I admit I'm really more of a software guy, but hard economic times means we gotta do thing outside of our realm of expertise, right?)
I don't have a great way to measure heat currently, but it seemed like Q8 was what was heating up. I had a probe from the original unit on with a 1M ohm resistor across it hooked up to the "to output" net-tag, which mirrored how the original unit was tested.
Using a finger is good enough. When a component exceeds 60 C it will be unpleasant.... That's good enough.
1 MOhm.... That's no load at all I would say.
How is that circuitry built? Are You sure there's no short caused by excessive solder? I used to inspect builds using a lupa....
It passed through our AOI without issue. We had it placed our our pick and place. It was pasted by hand with a prototype stencil since it's just a prototype at this point. It's definitely uncomfortable to touch, and it's the only part that was getting hot on the board.
AOI is automatic optical inspection. It allows us to check for solder shorts, missing parts, and such.
As far as the second schematic goes, PWM1 and 2 go to an output pin on the arduino, I have code switching between the two (with a small delay in between), so that the output switches at the desired frequency. So when PWM1 is high, V+ is passed through to the "pwm in" net-tag on the first schematic allowing 75V to go to the output. The same happens with PWM2 passing V- through causing the -75V to be sent to the output.
