Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations IDS on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Guidance required for 1phase/3phase Static Switch design 1

Status
Not open for further replies.

Gokkul55

Computer
Mar 1, 2013
44
Dear Members

I am currently designing a static switch driver circuit (for thyristor based switch) for 1phase and 3phase AC supply. The same circuit will work for 1phase if jumper is not set and will work for 3phase if jumper is set.

The help i need is in switching off the thyristor. There are 6 thyristors 3 for primary source and 3 for secondary source

primary source -- UPS with battery backup

Secondary source -- Grid

This static switch is the device that will be placed inside the UPS.

Consider 1Phase operation.

Here i switch on the primary thyristor (only one - as it is 1phase), and check zero crossing of primary supply to sense supply failure. On supply failure i switch will switch on the secondary supply.
But the switch being a thyristor it will switch off only on current being zero, and this equipment will be used in industrial area so the load would mostly be inductive can also be capacitive. So the load current can lead or lag the AC voltage.

My plan right now is to insert a delay between the two thyristor switching (delay equal to about 300 degree in angle). Is there any other reliable method to safely switch thyristors. I want the transfer time to be minimum as possible, so that computers wont reset.



Once 1phase is done, need to do the same with 3 phase.

Help appreciated

Thanks
Regards
Gokkul
 
Replies continue below

Recommended for you

You will need two thyritors to switch single phase AC. Otherwise, you get half-wave rectified DC. And that is very bad idea.

If you are talking low currents, like a few ten amperes, you could use triacs instead. There, you only need one for single-phase.

Gunnar Englund
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.
 
Skogsgurra said:
You will need two thyritors to switch single phase AC.

Sorry i forgot to mention it is two thyristor (part no - SKKT-92)

Skogsgurra said:
If you are talking low currents

No not low current - 50A minimum, for higher current the switching device will be upgraded but not the driver. That is the plan :)

Thanks
Regards
Gokkul
 
Been doing some and reserch and found a app note by nxp for detecting the triac off state by monitoring gate voltage level. Attached the app note for ref.

Also wanted to know if i keep the gate continuously on, read some where on the internet that it would create some gate loss of some kind and it is not recommended, not sure about it can some one enlighten me :).



Thanks
Regards
Gokkul
 
 http://files.engineering.com/getfile.aspx?folder=5c6c8923-51bc-4856-93e7-8d90d70c2e5f&file=AN467.pdf
Yes, monitoring the gate voltage is one way to detect current. The gate PN junction reflects voltage drop when current is flowing and that can be used to detect current. Perhaps not the most suitable method in this case. Or in any case at all.

The voltage across the thyristor is a (usually) esier and less interference-prone method. A conducting thyristor shouldn't have more than a couple of volts across it. If voltage is more, then it conducts. And then, of course, there are many ways to detect current in itself.

I use this method in our "Square variac" to avoid shorting the secondary windings while changing output voltage.
Works very well.

Gating on continuously is not a good idea. If you do that, you will require heavuer gate drives (more power because of continuous gate curent instead of short pulses or burts). And keeping gate current on during reverse plarity increases losses tremendously. You may get away with it in you case because you don't haveany reverse voltage (other thyristor conduts), but I don't see why you should do it.


Gunnar Englund
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.
 

Will try doing some research on voltage across thyristor method to detect triac off.

So about the continuous gate on you mean it wont do much harm as i am using triac (as one thyristor will always be conduction).

Skogsgurra said:
but I don't see why you should do it.

Just to simplify the code in this project :). Kind of switch it on and forget, rather than generate pulse.


Thanks
Regards
Gokkul
 
The triac does only have one gate, as you know, and keeping it on usually will not harm. But I would't use a triac in a 50 A RMS application because the gate usually is less sensitive in QIII than in QI and you need to provide the higher current continuously instead of discharging a capacitor long enough to fire the circuit.

It is not very complicated to turn gate off after you have turned it on. You will not save much code there.


Gunnar Englund
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.
 
[quoteSkogsgurra ]But I would't use a triac in a 50 A RMS application [/quote]

What else do u suggest (IGBT?), Mosfets are not an option as the current will increases to 250A and above in higher rated UPS. This is just a driver part of much larger machine.

Thanks
Regards
Gokkul
 
Thyristors, as you said that you would: " Sorry i forgot to mention it is two thyristor (part no - SKKT-92)"

Gunnar Englund
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.
 
I had recently found a Chinese hardware used in switching Skkt-92, as a static switch, circuit and pic attached.
This same circuit is available in the app note-page 106 also attached.

The resistors R18. R19, R20, R21 and R22 are 2 watt i guess(guessing from size).

I have also attached screenshots of voltage across the triac terminals and the current in the circuit (yellow - voltage waveform , pink - current waveform)

How do i sense the voltage across triac terminals to be detected by PIC microcontroller (either as digital io, or as analoge value)

Can this driver be used to continuously drive the triac gate or is pulsed gate still recommended?

Help Appreciated :)

Thanks
Regards
Gokkul
 
 http://files.engineering.com/getfile.aspx?folder=ef52ea37-2780-4877-9291-2ceaa9a15091&file=Screen-Shots.zip
Status
Not open for further replies.

Part and Inventory Search

Sponsor