Skip to main content
Topic: The "Engineer's Assistant" probe (Read 14053 times) previous topic - next topic

Re: The "Engineer's Assistant" probe

Reply #30
hey!
did you eventually manage to take a look at my schematics? I really need a second opinion, anyone :)
Thank you, and excuse me for nagging!

Re: The "Engineer's Assistant" probe

Reply #31
I just took a look, but I'm sorry, the analog is a bit beyond me. Do you have a brief description of the circuit and function? I will let let you know if I follow, or if it seems crazy wrong :)
Got a question? Please ask in the forum for the fastest answers.

Re: The "Engineer's Assistant" probe

Reply #32
Wow, thank you Ian!
well, the general idea is this: to inject voltage on the lsp1 pin, I use the q3 mosfet (which acts like a variable resistor) that I control via PWM from the microcontroller, using C5 and R12 as low pass filter. The resistor R6 is there just to limit some current, just in case (I still have to chose an appropriate MOSFET).
The "sensing" part is made of an op-amp to have high-impedance, but it has controllable gain via the the q2 mosfet that acts (well, it should, i didn't simulate) The gain is controlled by putting a PWM signal on the gate of Q2 via a low pass filter. Does it make sense to you?

OT: What do you guys use to simulate analog circuitry? I'm on a Mac and can't find anything easy to use, there is ng-spice, but does anyone know where can I find all the models for op-amps, mosfets, etc. ? Thanks!


Re: The "Engineer's Assistant" probe

Reply #34
Thanks, arhi, I'l just have to reinstall windows in my bootcamp partition and reinstall ltspice or multisim :P

Re: The "Engineer's Assistant" probe

Reply #35
I use LTSPICE on occasion. I wouldn't call it 'easy to use', in that the interface is pretty awkward and non-standard :)

You description does make sense, though it is too analog for me to give you any useful feedback in that department. I would like to play with this type of circuit though, I think it would be handy in an eventual 3.3volt version of the transistor tester.

Do you plan to use hardware or software PWM? It looks like the (single?) hardware PWM is on pin 5/RC5.
Got a question? Please ask in the forum for the fastest answers.

Re: The "Engineer's Assistant" probe

Reply #36
Thank you ian!

Actually, that part of the schematic is incorrect! I just noticed :/ What was I thinking? :D well, I don't know why I was convinced I had 2 PWM outputs, so it looks like I'll have to do everything in software... Well, not a big deal, I think, I'll just keep the same frequency on both outputs.

Hmm.. what if I use one hardware PWM to control gain and software PWM for voltage injection?

Re: The "Engineer's Assistant" probe

Reply #37
If you are making one soft PWM, it's not such a big deal to track a few more :) I would go all soft or all hard unless there was one feature that could benefit most from the increased resolution and accuracy of hardware PWM.
Got a question? Please ask in the forum for the fastest answers.

 

Re: The "Engineer's Assistant" probe

Reply #38
ian, my thoughts exactly!
I need some more resolution for gain control, otherwise measurements will not be accurate. Also, gain control is always in use, whereas signal injection is used only in some modes ;)