# TRIAC keeps failing

IMPORTANT UPDATE (Update 6), scroll down to see.

I have been trying to make an AC (air conditioner) soft starter for my home AC unit.

I have built the following circuit and connected it to my AC unit compressor. As soon as the AC starts the TRIAC fails and becomes a short between T1 and T2 pins. I have tested the circuit with smaller loads like a small 1 HP air compressor and it worked fine.

Schematic:

Code:

https://www.codepile.net/pile/wXkBew0R

Any ideas on why the TRIAC keeps failing?

Update 1:

I forgot to mention that the AC unit has a big capacitor on it but I couldn't read it's value.

Update 2:

AC unit specifications: Phase 1, Volts 220-240V, FREQ 50Hz, Max Input 15.8A

Update 3:

Start-up waveform:

The sequence starts like this and the on time keeps getting longer until it gets to the end which is always on.

Green is the Zero cross signal from the optocoupler. Yellow is the TRIAC_PIN

Later on:

Update 4:

The 'Soft start' effect on the scope with an air compressor: The differential probe is on a 1 ohm shunt.

Without soft starter:

With the soft starter:

Current rises slowly over time.

Update 5:

Here's the current (yellow) and voltage (green) waveform on the compressor:

Update 6:

Okay, so I decided to scope the AC unit without the starter and results were quite different, have a look the following:

The current shunt is 0.1 ohm.

AC unit startup:

Normal Operation:

So, apparently the startup current is almost around 100A (70A RMS)! with 150 - 160 phase angle between current and voltage. Under normal operation it consumes 23A (16A RMS) but the current wiggles! have a look at this video.

Update 8:

I sacrificed another TRIAC to check it whit scope and see what is destroying it but, I couldn't find anything.

Current = yellow x 10, voltage = green x 10

• The long comment chain has exceeded what is reasonable for comments. Therefore it has been moved to chat and should be continued there (link below). --- As this bulk moving of comments to chat can only be done once, any further comments posted here which try to clarify and understand the question, might be deleted without notice. Keep it in chat, please! When someone has got enough information from the chat to post an answer, then please do that. Any factual updates to the question which are decided during the chat, should be made via an edit to the question. Thanks. Jun 18, 2021 at 17:55
• Comments are not for extended discussion; this conversation has been moved to chat. Jun 18, 2021 at 17:55

Sounds like the triac is underrated for the surge. There are somewhat larger triacs such as CMA80MT1600NHR, but good luck finding any in stock at the moment.

Two SCRs are also a possibility (and likely much more available in hefty module form), and you could trigger them with your single opto. See Jack Creasey's answer here

Your circuit is not a "soft starter". In other words "zero detection" is not a "soft starter"

Compressor inrush of 100+ amps sounds about right for a compressor with RLA of 15-16.

The compressor will have a large inrush current for at least 160ms... until the compressor has gotten up to speed.

You are exceeding the "inrush rating" of the BTA41. You can pound it for an full AC cycle or two. Not for 100+ cycles of 100 amps.

Soft starters for home AC units (for 2-5 ton condensers) cost $300-$500. They are designed to "learn" how to startup (using a current ramp map) and detect when the load has change. In other words they have a built in relay "just in case". When the present inrush load exceeds the current ramp map used by the processor in the soft starter it "bails out" to a trusty relay.

Soft starters I install need 4 or 5 compressor starts to learn/develop the current ramp map.

Complementing the answers from Spehro Pefhany and pianoman, here is the datasheet for your TRIAC (BTA41) from ST.

The relevant part is figure 5:

Note the two curves: the top one is for non-repetitive surges, with the internal chip temperature at 25°C; the bottom one is for repetitive surges with case temperature at 75°C. Your actual situation could be represented by a curve inbetween.

Anyway, assuming your TRIAC is at room temp, it could handle 100A peaks only for about 250 cycles (20 ms each), so about 5 seconds. Most probably you are in worse conditions, since the circuit is probably encased and mounted somewhere where even if just starting "cold" the real internal temperature could be higher than the 25°C of the top curve.