Ground plane isolation question

Click For Summary
SUMMARY

The discussion focuses on the proper isolation techniques for a digital board interfacing with a high-powered output stage generating square wave pulses up to 500V. Jason O initially considered using the IL710 isolator, which offers 2500V isolation. A key point raised was the use of a high-value resistor, such as 10 MegOhm, between the ground planes to prevent static voltage buildup without compromising isolation. The conversation also highlighted the importance of adhering to UL standards for AC Mains isolation, which dictate specific resistance and capacitance requirements.

PREREQUISITES
  • Understanding of galvanic isolation techniques
  • Familiarity with the IL710 isolator specifications
  • Knowledge of UL standards for AC Mains isolation
  • Experience with circuit design involving high-voltage components
NEXT STEPS
  • Research the specifications and applications of the IL710 isolator
  • Learn about the design and implementation of high-value bleed resistors in isolation circuits
  • Study UL standards for AC Mains isolation, focusing on resistance and capacitance requirements
  • Explore star ground configurations and their effectiveness in noise coupling prevention
USEFUL FOR

Electrical engineers, circuit designers, and anyone involved in designing isolated digital systems for high-voltage applications will benefit from this discussion.

Jdo300
Messages
548
Reaction score
5
Hello All,

I was talking to a friend about how to properly isolate a digital board from a high-powered output stage. The output stage in my application make square wave pulses that can be as high as 500V. So my first thought was to galvonically isolate the output stage circuitry from the digital board which contains a microcontroller to run it all.

Up to this point, I figured that a simple isolator (like the IL710) would work great. They are fast, and have 2500V of isolation between the input and output.

But my friend mentioned that there should be a 10 ohm resistor between the two ground planes of the digital and output stage board to keep a static voltage from building up and poping the isolator IC.

But I'm thinking that if you connect the ground planes at all, then it is kind of defeating the purpose of the isolation in the first place. But at the same time, I recognize the issue with the static voltage buildup. How do I deal with such a problem?

Thanks,
Jason O
 
Engineering news on Phys.org
The value is just off. He probably meant more like 10 MegOhm, or 1MegOhm would work too. Bleeding off static charge doesn't take a low value resistor.
 
Ok, that makes sense. Is this what is commonly done in isolation situations? Will having the high value resistor still block fast transients from crossing into the digital side and blowing stuff up? I do have a lot of bypass caps and TVSs on the chips to prevent this from happening but maybe I'm just being paranoid...
 
Depends on the isolation requirements (like, is it for UL approval of an AC Mains powered device, or isolation of analog and digital sections of a device for noise coupling issues, etc.).

A high-value bleed resistor is commonly used when you don't want static buildup between two things. Like between the shield and the twisted pair inside shielded twisted pair cable, for example.

Often, just using a star ground configuration, with the two grounds connected only at one point (where the power supply is, for example) is sufficient to prevent noise coupling between two parts of a circuit.

When you're following the UL rules for AC Mains isolation, there are spacing rules, and rules for the allowed resistance (big) and capacitance (very small) between primary and (SELV) secondary circuits.
 
Last edited:
Ahhh interesting :smile: Thanks for the info
 
Most likely this can only be answered by an "old timer". I am making measurements on an uA709 op amp (metal can). I would like to calculate the frequency rolloff curves (I can measure them). I assume the compensation is via the miller effect. To do the calculations I would need to know the gain of the transistors and the effective resistance seen at the compensation terminals, not including the values I put there. Anyone know those values?

Similar threads

  • · Replies 1 ·
Replies
1
Views
2K
Replies
8
Views
4K
  • · Replies 17 ·
Replies
17
Views
6K
  • · Replies 6 ·
Replies
6
Views
4K
  • · Replies 4 ·
Replies
4
Views
17K
Replies
11
Views
14K
  • · Replies 1 ·
Replies
1
Views
4K
Replies
11
Views
14K
  • · Replies 7 ·
Replies
7
Views
10K
  • · Replies 19 ·
Replies
19
Views
3K