advice about solid sate relays please 1

[les20]

I am using four SSR's to continually reverse the current through an inductive load. The load is in the form of an electromagnetic coil with a resistance of 2.5 ohms and about 1000 turns of 0.1mm dia. copper wire. I am driving the input to the SSR's with a TTL from an encoder. I am using a freewheel diode on each of the four SSR's to deal with the transient suppression. The frequency of operation is from zero to around 60Hz. The SSR's are rated at 100V output and 40A load. I have chosen the freewheel diodes with care but on occassion they fail and the 40 pound diode/s break down!! Any comments on dealing with high inductive loads would be helpful. Thanking you in advance for your help.

 

[Skogsgurra]

Hello les20,

I assume that you are using an H-bridge configuration. I also assume that you know about the "water hammer effect" when the reverse current in the diodes is interrupted. In your highly inductive circuit, it will cause a high kick-back voltage that kills your diodes.

A snubber parallel to each diode will help a lot. The snubber should consist of a resistor and a capacitor in series. The resistor value could be something like your load resistance or higher. I would chose 2.7 ohms. The capacitor shall be able to absorb the charge and 100 nanofarads is universally accepted as a working size for most snubbers. Make the resistor 2 - 5 W and see to it that the capacitor can take at least 200 V DC. You need a HF capacitor and keep the leads short.

 

[sreid]

1)Are the diodes "Ultrafast" rated? Reverse recovery time with slow diodes can cause faiures.

2) Are the diodes on heat sinks? Do they get "hot."

3) Are the leads to the diodes physically short to minimize inductance?

 

[les20]

Dear sreid,
Than you for your comments, the diodes are fast but I will check again to make sure they are super fast. The diodes are on heat sinks and the leads are short.Thank you again all comments like this are welcome.
Regards
Les

 

[les20]

Hi skogsqurra
Thank you again for your help, I have been thinking about it. Some qestions please.
Is it better to use the snubber with the present fast reaction diode than use a superfast diode if I can find one?
I'm not sure how the snubber works to improve the diode. I have tried to work it out and have this solution-is it correct?
The TC of the snubber works out to 27 micro seconds. The max. turn on and turn off time is for the D1D40 is one msec. Is it that the discharge of the capacitor when the SSR is turned off, which is in fact quicker than the turn off time, in some way protects the SSR output switch?
Thank you for your help.
Les

 

[Skogsgurra]

les20,

Have a look at this pdf:

http://www.irf.com/technical-info/whitepaper/murdiodes.pdf

It describes the turn-off transient and also gives some rules for optimum snubber design. You may find that my chosen values are quite far from optimum, but they are a lot better than no snubber at all.

 

[les20]

Hi Skogsgurra
Thank you again, I will study the link that you suggest.
Les