PWM Control for 3 HP DC Motor: Snubber Help Needed

Too much energy is dissipated in the diode alone.

You need to add an appropriately sized/rated R and C. The simplified explanation is that the diode directs the excess energy (inductive switching transient) to be stored in the capacitor. The R dissipates this energy during the interval between switching events. If the RC+diode snubber is correctly implemented, all parts should last for years.

I could provide more details and/or design the snubbers for you, but my employee handbook says ALL my ideas & designs (24/7/365) are owned by The Company. You would have to negotiate compensation with Them ;-)

Good luck!

Perhaps some help with formulas to produce my own answers? I am old and feeble and need all the help I can find.. Thanks for your input.. John

Sorry link in first post didn't work. Try these...

http://en.wikipedia.org/wiki/Snubber
http://ken-gilbert.com/images/pdf/snubber.pdf

or search Google for "rcd snubber design formulas".

Best wishes.

PWM at 88Hz? Did someone with any idea on electronics designed a PWM at 88HZ? you propubly need an external coil larger than the motor for this to work. And used a 1200V IGBT? On what operating voltage? The higher the IGBT sustaining voltage, the larger the Vforward and higher the conduction losses all other things being equal. S.M.

The motor runs at 300 VDC and has a substantial armature winding, 6 ohms dc.. 40 amp spikes at startup.. Yes, I experimented on a 1/4 HP motor and at higher frequencies losses are very apparent. Normally this motor is best suited with SCR phase angle control but due to the facilities phase imbalance we thought to give PWM a go (no off the shelf drives available with the desired features that could withstand the phase imbalance.) Yes, I am a little out of my forte here, as stated, I have evolved into a programmer but the project is intriguing and has forced me to use some of that schooling from 30 years ago.. Maybe a little too much has been lost over the years... Thanks for the help..

My initial guess would be to use a Schottky diode instead of a fast recovery diode. The voltage drop will be lower and thus the energy dissipated by the diode will be less. This will be valid only if the diode steers this returning energy into circuitry to recover this back EMF energy. If the momentary 3 HP of back EMF power (I know, a physics oxymoron) has only wire resistance and diode drop to dissipate these Joules then you will continue to pop these diodes or something else.

don't forget that a electric engine (any) when is started will consume about 6 times the current from is nominal value, then will happen a different response from a) the test engine, and b) the load engine. that freq. seams funny... is it the main carrier frequency or the PWM modulated pulse frequency?

Thank you, very good information. I had used an FOD3180 for the driver but see your point, I will work to make these changes.. I still need to resolve my snubber diode issue, I had actually used a 30 amp rated diode but when I reexamine data I see that I may easily see back emf spikes in the 50 amp range when applied across a diode only; I am thinking to just select a new diode, one that will handle higher current.. If you have any input here, it would also be appreciated.. Thank You. John

When the fast recovery diode conducts it momentarily raises the DC Bus voltage way above the safe limit for your IGBTs - even if they are rated at 1200V and running off 300 VDC. You need to add enough Bus capacitance to absorb the back EMF.

Motor controllers always have heaps of capacitors built in for this purpose - as do 3 phase inverters (VFDs).

For 3HP off 300 VDC I would suggest around 1000 to 2000uF if using low ESR electrolytics. 100 to 200uF for Film.