500 VAC Isolation

Thread Starter

Dave MacK

Joined Jan 9, 2012
6
Hello,

Me and my partner are working on a project with PIC micro controller. The project requires us to take a 500V pulse from a device and to read it with the PICs analog to digital converter. The problem we are having is that the 500V to 5V transformers we have found are either too expensive or unavailable.

Is there any suggestions here from people on how me and my partner can cost effectively isolate this 500V source safely from our PIC, perhaps using an opto isolator?. Also, the voltage we are trying to read is a very fast pulse, so that must be accounted for.

Thanks!

-Dave
 

t06afre

Joined May 11, 2009
5,934
Can you tell us more about the pulse. Like any timeing details or anything alse that matter. Do you have a scope picture of that puls. That would also help.
 

t06afre

Joined May 11, 2009
5,934
What is this school project about. Is it just pulse counting, or something more advanced. Also which PIC will you use
 

Thread Starter

Dave MacK

Joined Jan 9, 2012
6
The project is as simple as measuring the peak voltage (500V) on the pulse. We dont have to worry about the frequency or wave length at all. Were using the 16F887.
 

kubeek

Joined Sep 20, 2005
5,794
Does it really need to be isolated? The signal is pretty fast, so the transformer design would not be easy to do right to make the slew rate fast enough so that the pulse is right. So if you don´t need galvanic isolation, then a simple resistive 100:1 divider should work nicely.
 

Thread Starter

Dave MacK

Joined Jan 9, 2012
6
My instructor requested we isolate it, he suggested we use opto isolation or a transformer. We were orignally just going to use the 100:1 Voltage divider itself, but he suggested we use isolation for safety. Im going to have a discussion with him and tell him what we read here. Hopefully he allows us to continue with just the divider. The cheapest transformers we could find were worth hundreds of dollars.
 

kubeek

Joined Sep 20, 2005
5,794
The cheapest transformers we could find were worth hundreds of dollars.
Exactly, accurate measuring transformers will cost A LOT. There are some optocouplers that can provide feedback and thus can be used for precise measurements, but I think they cost a lot too. How much precision do you actually need?
Also, you could do the A/D conversion on the 500V side and send the digital data over optocouplers, which might be easier.
 
Top