terrible spikes with PWM mode

General discussion of the MCP motion controller product line
Post Reply
fausto.tromba
Posts: 37
Joined: Wed Jul 31, 2019 7:24 am

terrible spikes with PWM mode

Post by fausto.tromba » Mon Jan 11, 2021 12:44 am

I have taken a new MCP236 and I wanted to check the voltage.
This because last time I found the current quite noisy.

Checking the PWM1 (DUTY 50 on 1000) connected to a Power resistance, I got unexpected voltage impulse every almost 15s with a width of 200ms.
(see LeCroy56.jpg)
When I connect the motor this impulse has an effect on the current (yellow track).

To avoid any doubts, I have measured also the power supply and it is stable at 48V.
LeCroy55.jpg

We get stuck since weeks
Attachments
LeCroy55.jpg
LeCroy55.jpg (79.9 KiB) Viewed 2074 times
LeCroy56.jpg
LeCroy56.jpg (58.86 KiB) Viewed 2074 times

User avatar
Basicmicro Support
Posts: 1554
Joined: Thu Feb 26, 2015 9:45 pm

Re: terrible spikes with PWM mode

Post by Basicmicro Support » Tue Jan 19, 2021 9:39 am

The scope shots you show show the noise generated by the PWM rising and falling edges. Can you show scope shots of the 15s and 200ms spikes you talk about?

If I understand you, all you did was set Duty to 50 in motion studio?

SimonRafferty
Posts: 7
Joined: Sun Sep 06, 2020 9:02 am

Re: terrible spikes with PWM mode

Post by SimonRafferty » Sun Jan 31, 2021 6:28 am

>Can you show scope shots of the 15s and 200ms spikes you talk about?
I think they already have!
If you look at the timebase on the scope, it says 40.8S. The repitition rate of the pulses is about 15S on that basis.

User avatar
Basicmicro Support
Posts: 1554
Joined: Thu Feb 26, 2015 9:45 pm

Re: terrible spikes with PWM mode

Post by Basicmicro Support » Thu Feb 18, 2021 10:47 am

They look like that because you have too much inductance in your ground lead. To get a clean look at inductive spikes you need to use the ground loop at the tip of the probe and keep the physical distance between the measurement points close together. The ground lead on most probes are fine for slow moving signals but the inductive spike is very quick and bounces in the probe due to the ground lead inductance.

Post Reply