0
\$\begingroup\$

We're having some trouble with a small inverter motor design (24V / 20A).

We're seeing quite a lot of ringing on the high-side mosfets, we scope it with the negative of the oscilloscope on the phase output and the positive at the +VDC and here's what we see :

Vds_HS

We're seeing spikes of over 8V over the VDC_bus and 7V below.

We're using a DRV832x for gate-driving and this is with the standard settings (1A sourcing, 2A sinking or is it the other way around?). We tried to lower these gate current values all the way down to 20mA but it actually doesn't change much the ringing, only makes the switching slower and sometimes even drops some events, maybe an internal protection against switching too slow?

enter image description here

We don't see this behaviour on the low-side at all where all is fine and dandy.

The layout is done according to the rules with trace lengths being balanced (we're driving 2 MOSFET per switch) and the inductive loops / stray inductance minimized as much as possible so we're quite surprised to see this behaviour. The DC link is basically 220µF of MLCC which should give us some low ESR / ESL.

Any lead on how to minimize this? Could it be a measurement effect of us not scoping it correctly?

\$\endgroup\$
5
  • \$\begingroup\$ We can't troubleshoot circuits we can't see. \$\endgroup\$
    – DKNguyen
    Dec 7, 2020 at 16:43
  • 2
    \$\begingroup\$ Are you sure this isn't just scope probe ringing? \$\endgroup\$
    – user_1818839
    Dec 7, 2020 at 16:48
  • \$\begingroup\$ Are you using the probe in 1x or 10x mode? What is the estimated ringing frequency? Is there a gate resistor to slow down the switching? Are you sure that 50MHz scope is enough to give any sane results? \$\endgroup\$
    – Justme
    Dec 7, 2020 at 16:52
  • \$\begingroup\$ Is the drive controlling the motor successfully? Are you having any component failures (mosfet or gate driver)? \$\endgroup\$
    – mkeith
    Dec 7, 2020 at 17:13
  • \$\begingroup\$ Hi all, In order : it might indeed be probe ringing. The probe is being used in 1x mode. The ringing frequency is in the 100's of khz but no precise value yet. The drive is controlling the motor perfectly and we haven't seen any failures so far... \$\endgroup\$
    – VoltsAndNuts
    Dec 8, 2020 at 0:01

1 Answer 1

Reset to default
2
\$\begingroup\$

We're seeing quite a lot of ringing on the high-side mosfets, we scope it with the negative of the oscilloscope on the phase output and the positive at the +VDC

That'll be your problem - the DS1054 doesn't have differential inputs and you are connecting the "grounded scope input" to the hot output line in your push-pull converter. Thus you are adding a whole lot of ground to the hot output line and usually that means shorting things out. However, in your situation, I expect the output is floating but, you are still adding nano farads of capacitance to a fast switching line and causing parasitic circuit inductances to ring.

Never do this unless you have proper differential input probes and, at $400 dollars, the DS1054 is unlikely to have them.

Could it be a measurement effect of us not scoping it correctly?

I believe so.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.