How to capture voltage waveform across triac pins

Thread Starter

noweare

Joined Jun 30, 2017
115
Schematic_Spot-Welder.pngHello friends,

I am working on a spot welder for making lithium ion battery
packs for my projects.

The triac gets destroyed whenever I try to weld with it. The two main AC pins
on the traic become internally shorted. When I use a 40 watt incandescent
light (instead of the welding transformer) it all works as planned.
Nothing blows up.

I am thinking it might be inductive spikes across the triac that is
blowing it up.

Another problem is that I cant seem to capture the voltage
across the triac pins to see if it an inductive kickback from the
transformer.

To try to troubleshoot the circuit, the mains voltage is replaced
with a isolated AC 24volt source but I leave the welding transformer in place
as the loac.

I have a 2 channel storage scope both probes are used to measure
the difference between main triac pins. Both probes have an AC signal
such that I can't seem to use either as a trigger.

This leaves me with out a probe to use as a trigger.

Anyone have any ideas on how I can try to capture the voltage across the
triac ?
 

danadak

Joined Mar 10, 2018
4,057
Sounds lkike you are doing the right thing doing a differential measurement.


Sounds like you trigger setup is suspect. What is model # of scope you have ?


Regards, Dana.
 

Thread Starter

noweare

Joined Jun 30, 2017
115
Hello Dana,
I just purchased this scope it is a Siglent 1202X-E . Just got it last week. It 's a 2 channel 200 MHz digital storage scope. Has a
barn load of bells and whistles that I don't know how to use. So learning it will be a project in itself.

I figured out to use the external trigger input to the scope. I had a extra scope probe and connected it to the output that triggers the optical diac that in-turn triggers the triac. So I was able to see the voltage differential across the triac pins during conduction. I had to subtract the channels since I can not use the ground lead.

I did not see any inductive spiking but then again I am only driving the transformer with 24 vrms compared to 120 vrms (real world). Now I am thinking it may be the di/dt rating I am exceeding since I only have 1 1/2 turns on the secondary.

I thought this would not take so long to build but proving to be not-so straight forward : (
 
Last edited:

TeeKay6

Joined Apr 20, 2019
573
View attachment 191245Hello friends,

I am working on a spot welder for making lithium ion battery
packs for my projects.

The triac gets destroyed whenever I try to weld with it. The two main AC pins
on the traic become internally shorted. When I use a 40 watt incandescent
light (instead of the welding transformer) it all works as planned.
Nothing blows up.

I am thinking it might be inductive spikes across the triac that is
blowing it up.

Another problem is that I cant seem to capture the voltage
across the triac pins to see if it an inductive kickback from the
transformer.

To try to troubleshoot the circuit, the mains voltage is replaced
with a isolated AC 24volt source but I leave the welding transformer in place
as the loac.

I have a 2 channel storage scope both probes are used to measure
the difference between main triac pins. Both probes have an AC signal
such that I can't seem to use either as a trigger.

This leaves me with out a probe to use as a trigger.

Anyone have any ideas on how I can try to capture the voltage across the
triac ?
As a spot welder your schematic makes no sense. The only power source shown is 120VAC at P1. P1-pin2 carries an AC signal to U1 for sensing zero crossing. P1-pin1 completes the zero crossing detection circuit. P1-pin1 is the only power connection to P2. For the case when T2 is off, follow your circuit from P1-pin1 through the triac circuit back to P2-pin2. When the triac is on, P2-pin1 and P2-pin2 are effectively shorted together...doing nothing. Where is power to drive the primary of T1 supposed to come from?
 

Sensacell

Joined Jun 19, 2012
3,432
When are you triggering the Triac?
What part of the cycle?

Most scopes have an external trigger input, some even have an AC LINE trigger option built-in.
Use the external trigger input and the two input channels are free for differential measurements.
 
Last edited:

LesJones

Joined Jan 8, 2017
4,174
This does not answer your question but I can see two things that are wrong.
1- I assume R6 and C1 are supposed to be a snubber. If so then the value of R6 is too high. I would suggest something like 22 to 100 ohms to be more suitable.
2 I think you will be exceeding the current rating of 4 amps of the triac by a large amount.

Les.
 

Thread Starter

noweare

Joined Jun 30, 2017
115
As a spot welder your schematic makes no sense. The only power source shown is 120VAC at P1. P1-pin2 carries an AC signal to U1 for sensing zero crossing. P1-pin1 completes the zero crossing detection circuit. P1-pin1 is the only power connection to P2. For the case when T2 is off, follow your circuit from P1-pin1 through the triac circuit back to P2-pin2. When the triac is on, P2-pin1 and P2-pin2 are effectively shorted together...doing nothing. Where is power to drive the primary of T1 supposed to come from?
You are correct the schematic is not right. In reality, the top of the triac is connected to Pin 2 of the top connector not Pin 1 of the lower connector. Good catch. Sorry 'bout that and thank you for taking the time to go through the schematic.
 

Thread Starter

noweare

Joined Jun 30, 2017
115
When are you triggering the Triac?
What part of the cycle?

Most scopes have an external trigger input, some even have an AC LINE trigger option built-in.
Use the external trigger input and the two input channels are free for differential measurements.
The software prompts for a delay that is input to an ADC from R9. The delay can be from 0 to 7 msecs.
So the triac is triggered on both the positive half and negative half cycles. The delay starts after a zero crossing is
detected by an interrupt. Also number of cycles is input using R9 which can be from 1 to 40 cycles. It's just to tune
the input to the welding transformer to give a good weld. (when I get it working).
 

Thread Starter

noweare

Joined Jun 30, 2017
115
This does not answer your question but I can see two things that are wrong.
1- I assume R6 and C1 are supposed to be a snubber. If so then the value of R6 is too high. I would suggest something like 22 to 100 ohms to be more suitable.
2 I think you will be exceeding the current rating of 4 amps of the triac by a large amount.

Les.
I think I am being too sloppy with the schematics . That resistor value is actually 47 Ohms, 2 watts.

I don't really know what I need for secondary current for a good weld.
The data sheet says max non-repetitive peak on-state current is 25 amps. So if I use half of that value
then my secondary current would be 400 amps or so (3.6 vrms on the secondary unloaded). If its more then that could be the reason why the triac is failing and like you said I need to go with a higher current part.
 

MisterBill2

Joined Jan 23, 2018
18,167
I see not one word about what the transformer is, and what it's no load current is specified as. I am amazed that nobody else really addressed that lack of information. First, the full load current from the transformer must be less than the steady-state rating of the triac, and second, the inrush current of the transformer must be less than the peak current rating of the triac. Until both of those conditions are satisfied, of course things will burn up. If that transformer is able to deliver 400 amps at 3.6 volts that equates to an output power of close to 1500 watts, and assuming 100% transformer efficiency, an input power of 1500 watts., over 10 amps at 120 volts. That would pop a 4 amp triac every time. That is a big part of your problem.
 

Thread Starter

noweare

Joined Jun 30, 2017
115
Hello MisterBill2,
The transformer is a modified microwave oven transformer (120vrms) . The secondary is 3 turns. I ended up using a 40 amp triac (400 amp pk non-repeatable) with a heatsink. You are correct about the inrush must be less than peak and I am assuming that inrush should be measured when the load is first turned on. I have not measured that. I would have to get either a current transformer or a shunt resistor to measure that. But it would be good to know to properly size the triac. I need to learn how to measure things accurately, the things that you mention in your post.
Thanks for your input.
 

MisterBill2

Joined Jan 23, 2018
18,167
Because the initial surge is fast and may even include ringing, you will need to have a rather fast transformer to capture the current wave form. Probably a fairly small ferrite toroid core with just a few turns of wire would be adequate, and less expensive than a suitable series resistor. It will also provide a much safer connection to the scope input since it would have no AC mains connection. Microwave power transformers have a magnetic shunt that you need to consider in the welding transformer design, although you may already be quite aware of that.
 

Thread Starter

noweare

Joined Jun 30, 2017
115
I have never tried making a current transformer. Ferrite toroids are fairly inexpensive. I would have to see what flux density
10 amps - turns is and see what size core I would need. I assume I don't want to be to high in flux density so the core losses
are low. Good idea.
 
Top