Pulse Transformer Design.

[deepakchikane]

Hi looking out to design a pulse transformer for the thyristors.?

Normally operating frequency will be 50 hz,current=50 ma with 1:1 ratio.

it will be driving through a controller.

i am very new in the transformer design side.
though i dont have idea how to select core.?

Kindly let me know the criteria to design such.

Look forward to know from you.

 

[FvM]

Did you consider to review datasheets of industry standard trigger transformers to get an idea of a usual specification? **broken link removed**
http://www.hammondmfg.com/pdf/5c0064-65.pdf

 

[D.A.(Tony)Stewart]

In the 60's we used to make our own Pulse transformers for Zero crossing or Phase control using short soft iron 1mm wires from small spool in hardware store ( like twist in parallel, then mylar or plumber's teflon tape then primary one end , secondary other end using 30 AWG wire or higher.

Logic Gate drive current only needs to exceed Triac Gate current minimum for 1:1.

ZCS can be designed easily or phase control.

Can be made quite small then epoxy or silicone dipped.

Check polarity to ensure Quadrant I or best quadrant and add protection as required with diodes, MOV's

 

[deepakchikane]

Did you consider to review datasheets of industry standard trigger transformers to get an idea of a usual specification? **broken link removed**
http://www.hammondmfg.com/pdf/5c0064-65.pdf

Hi FVM,

There are lots of TRANSFORMERS listed here with different inductance range.
what is the selection criteria..??
1) Primary voltage pulses from controller=5v pulses
2) Frequency=50hz for triac triggering
3) current requirement will be 50 ma max

i calculated the inductance.
V=L(dI/dT)
v=5v;
di=50ma
dt=20ms

it is coming almost several mH.

Is there any other way to calculate..?
the same mH are not listed in the datasheet.

Kindly help to sort out the issues.

- - - Updated - - -

Hi sunny,,

Whats the polarity..?
i gone thru different articles.
its seem to be a anti polarity configuration.

kindly help with the correct polarities.

 

[FvM]

Did you look at the µVs Voltage Time Integral parameter in the datasheet? It clarifies that the trigger pulse can't be longer than a few 10 up to 100 µs. That's no problem for the thyristor, but your trigger generator has to be designed respectively.

 

[deepakchikane]

FvM
Again, whats the slection criteria.
i am using Triac ACST12357.

Where now where mention the uVS time integral parameter in the datasheet.

will i select the the max trigger pulse 100 us for the better performance or 10 uS..??

 

[FvM]

You'll set the trigger pulse width according to the thyristor/triac requirements. Most devices turn on within a few µs with sufficient gate current.

 

[deepakchikane]

You'll set the trigger pulse width according to the thyristor/triac requirements. Most devices turn on within a few µs with sufficient gate current.

FvM
Then this will not operate on 50 hz signal.

i think mostly i will have go for higher frequency with 10% rise time.. is that right..?

 

[FvM]

You didn't tell yet what your application is, in so far it's not clear what you mean with "50 Hz signal". If your application requires for some reason to trigger a thyristor or triac permanently over a longer timer period, e.g. because the load phase angle is unknown, you'll usually go for a burst of short trigger pulses.

 

[deepakchikane]

Hi FvM,

i am making a static voltage regulator where i am replacing opto coupler with pulse transformers.

initially i read some article to fire a triac so i thought i will have fire a triac at 50hz signal.

now looking at the size of readymade pulse transformer i think i can fire the triac with the high frequency by keeping lowest o time.

is anything needed..??

so 100 khz signal from controller with 10 % duty cycle will give 1 us pulses so i can trigeer the triac with pulse transformer
correct if i am wrong?

 

[FvM]

Sounds basically good. I would prefer a pulse width of 5 or 10 µs and respectively lower pulse frequency. Higher duty cycle won't be a problem with a circuit as suggested in post #3, but 10 % is a good starting point to save power.

 

[schmitt trigger]

One thing that has not been discussed is the isolation and safety requirements. These will dictate your actual construction methods and materials to be used.

Once that you have your first samples, you must test the dielectric withstanding voltage, to the voltage level required by the safety agencies, which vary by country.

 

[Anna Conda]

So called picket fence triggering, 50% pwm, at 50-100kHz, using a small ferrite Tx, works well, unipolar pulses out of the Tx via one diode, can keep this gate drive going for at least 50% of the desired on time of the SCR to ensure turn on...

 

[D.A.(Tony)Stewart]

Since fences are exposed to more lightning spike levels than telephone lines, consider a spark plug arrestor gapped to 1kV/mm and fuse where necessary with 25kV rated fuse. I.e. at least 25mm long.

 

[FvM]

Picket fence triggering is meaned figuratively, I presume.

In older thyristor controllers, you often find an unijunction transistor pulse generator driving the trigger transformer.

 

[D.A.(Tony)Stewart]

Deepak, can you define end result? Phase control? Linear? Logarithmic? Speed control? What load? Or just a big sawtooth?