Ssr Q ?

[andy55]

I need a solid state relay for switching a 24 V DC for the brake on a servo motor.

I have looked at various options and the cheap small SSRs all seem to be of triac type mainly intended for switching AC mains voltage.

My question is, can this triac type of SSR be used for switching on and off a simple DC voltage also ??

In particular I'm looking at the sharp DIP8 packaged SSRs, for example
http://document.sharpsma.com/files/pr3bmf11_e.pdf
they are about 1.20eur from digikey and would be ideal for me if they can also be used for switching a DC voltage/current.

TIA

[gar]

051212-0950 EST USA


andy55:

Any Triac or SCR device after being triggered will remain on until the current thru the device drops below a holding current which is relatively small compared with the maximum current rating of the device.

In an AC circuit this happens once every half cycle.

In a normal DC circuit this never happens until you use some other means to force the current to less than the holding current.

To switch a DC circuit you need to use a mechanical contact, a transistor, FET, or if you use an SCR or Triac, then you need some other means to force the current low such as a capacitor and another SCR.

Any time you switch an inductive load you need something to limit the peak voltage from the inductive kick. Usually for your type of load you would put a reverse biased diode across the inductive load unless drop-out time was critical, then an RC snubber would be used.

If you search far enough you can probably find an optically isolated, or some other kind of isolation, SSR based on an FET.

How often is this brake operated? What is wrong with a P&B KUP relay? For 35 years we have used the KUP to switch large clutches ( 120 VDC at 1 A ) in automotive pinion preload adjust machines. Here the clutch is switched on and off several times every minute 24 hrs per day. Typical life about 6 months. If you used another relay to periodically change the direction of current flow thru the controlling relay then the contact life could be considerably extended. The type of failure we encounter is a unidirectional metal transfer due to the DC that produces a conical metal mound on one contact and a mating hole on the other. Ultimately the contacts weld together. We use the KUP 11D15 which has silver cadmium oxide contacts that are better for inductive load circuits at 120 VAC, but not good for low voltage, 5 v and below.

Another possibility is to use the SCR to turn on the brake, then short the SCR with a mechanical relay contact, remove the SCR trigger, and open the mechanical contact when you want to turn off the brake. Turn on with the SCR eliminates problems from mechanical bounce of closing the mechanical contact.

.



.

[andy55]

thanks for the explanation.

better stay away from the triac types then...

couldn't find something similar (small, cheap) for DC that easily though...

I would prefer a SSR over a conventional relay since that saves me two or more components (with conventional relay: optoisolator + transistor for switching coil)

[gar]

051212-1236 EST USA

andy55:

If you use an ordinary electromechanical relay it is an isolator.

Describe your load --- voltage and current. Is the brake enabled on loss of electrical excitation, or vise versa.

.

[andy55]

Originally Posted by gar 051212-1236 EST USA andy55: If you use an ordinary electromechanical relay it is an isolator. Describe your load --- voltage and current. Is the brake enabled on loss of electrical excitation, or vise versa. .

This is for the brake on a sanyo denki p5 servo motor. it is a solenoid which requires 24 V over it to release the brake.
I don't have the specs right now but the current was maybe 250 mA or so.

I want to drive this with TTL 5 V signal from control logic.

Yes you are correct no optoisolator needed with relay. Just a transistor to switch on/off the coil of the relay.

[gar]

051212-1326 ESTUSA

andy55:

Here is another way:

Put an isolated Triac type SSR in the 120 primary side of a transformer, a bridge rectifier on the secondary to supply the brake, no filter capacitor, and a reversed biased diode across the brake.

The average DC output voltage without a capacitor is 0.636/0.707 times the AC sine wave RMS voltage.

No filter capacitor will minimize the decay time which will be set by the brake coil and back biased diode.

.

[Al_The_Man]

Or what you could do is put the Triac in the ac side of the 24v supply to a bridge rectifier , switch the triac with opto coupler like Motorola MOC3020, and connect the brake directly to the bridge DC output.

[gar]

051213-1108 EST USA

andy55:

I just ran a test on our large clutch with a shunt reversed biased diode, and no added resistance.

Note: Electro-mechanical clutches and brakes are essentially the same device, except for whether you close (brake) or open on loss of power.

Our large clutch is modified to have a larger than normal air gap to reduce drop out time. When used with an RC snubber the drop out time is about 75 ms with a torque load of about 200 #-ft. This clutch slips at about 500 to 700 #-ft with the large air gap and full excitation. The peak reverse voltage with the snubber is about 1500 v.

With an MR752 diode and no added series resistance across the clutch coil and no torque load the drop out time is about 2 to 3 seconds when excited with 108 vdc.

If you use the Triac bridge rectifier technique, then use a high PIV bridge and add resistance in series with the shunt diode. Choose the resistance to limit the peak voltage from the coil to about 1/2 the PIV rating of the bridge. Start with a low resistance. As a guess if your coil current is 0.25 amps, then peak voltage is = 0.25 x R. From a power standpoint you can probably overload a wire wound resistor 10 x for a couple of seconds. For 200 v peak at 0.25 amp the resistance is 800 ohms. Continuous power is 200 x 0.25 = 50 watts. I would use a 5 w resistor. 2 w would probably work if the decay time is short enough.

.