[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [TCML] Sidactors in a can of worms



Hi Finn, James,

These may be specific for telecom with something different internally. Thanks for contacting Littlefuse. Maybe they can lead us to an equivalent switching sidac that is not obsolete or heading that direction. Value isn't overly important (just series them for standoff). It may be hard to find a direct replacement at a specific voltage or even near. We can always go to TO-92 or even DO-15's. As a matter of fact, series two K1500G's (and Mouser has 1200 in stock at 0.56ea in qty of 1) in place of the K3000F1's would likely do the job. The sidac's are staying cool considering the nanosecond on and off times of the IGBT takeover, so I really don't see a problem with that (and can be used with existing boards) ..... and should act like the Sidac's we love.

Another option (for future thoughts) is using a Quadrac. The internal diac (gate pin) would need a cap and resistor to set the breakover voltage, but this should be easy enough to do and would lend itself to triggering without the need of trigger coils? ;-) .

Well, I'll be interested to hear what Littlefuse has to say. Thanks so much Finn for performing this testing. This is exactly "why" testing is very much needed.

Take care,
Bart

Finn Hammer wrote:
I got the Sidactors yesterday, Thanx Terry, and immediately took action to measure their break over voltage. Something soon seemed odd, they looked like TVS'es to me, not sidactors at all.

When I applied voltage across them from my current limiting bench supply, a HP6255a, the devices did respond to break over voltage, but failed to break over to a near short, as advertised by Teccor, and as previously experienced by the KxxxxF1 series that we have been using up to now. What I saw was a device that plainly clamped the voltage to the break over level, and kept it there.

This is the behavior that I`d expect from a Zener and a TVZ, but not from a Sidac.

Of course, there might be a limitation of my measuring approach that I had overlooked, holding current came to mind, so I decided to lash up a birds nest SISG, to test out the new types against a known good one, the K2500F1, which has shown a consistent break over voltage of 135V per section.

The K2500F1 single section worked as expected and produced this oscilloscope trace:
http://www.hammertone.com/temp/sidac.jpg
In this screen,
http://www.hammertone.com/temp/sidac.jpg
Top trace, cyan CH2, is gate voltage, 15.2V
This is a bit low since generated from a 24V TVZ, but the current is low too, as seen on 3rd trace.

2nd. trace, magenta CH3, is current into gate, measured with Tektronix P6022 current clamp and passive terminator.
1.36A into gate.

3rd. trace, green CH4, is current in tank circuit, 17.8AP-P measured with Pearson Current Monitor #101

4th trace, Yellow CH1, is voltage across tank cap., 135V
Screen says 62V but probe is MX9003 differential by Metrix in 200:1

So far so good.
But when I plug any of the following sidactors into this circuit, all I can record is that the voltage across the cap rises to, and stays at, break over voltage of device. What I don`t see is any rapid discharge of tank capacitor, no voltage rise on, or current into gate, and thus no oscillations. So as far as I can see, these Sidactors are not what they are advertised to be, but instead TVS, Zener whatever derivatives unsuited for SISG duty.

I`m at my wits end with this, because -frankly- I feel out of my depth claiming this in the face of a major manufacturer like Teccor.

I guess I's time to write them, but I'd just bring it up here first.

Cheers, Finn Hammer

_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla

_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla