Heat is the reason I'm actually doing this. I got some 8amp SSR's but with 5 amps it starts getting very hot so no way I'll get to 8 amps without a couple pounds of metal. So my thought is get really high current triacs so maybe they don't get as warm with 7-8 amps. I'm getting 3 different ones, 16A, 20A and 30A and getting a few different heat sinks just in case. Ultimately I would love to have no heatsink but that might not be possible? What's your thoughts about this approach?

https://www.digikey.ca/product-detail/en/T1620T-8G-TR/497-18458-1-ND/9866952/?itemSeq=290572499

https://www.digikey.ca/product-detail/en/T2050H-6G/497-17284-ND/7313725/?itemSeq=290473809

https://www.digikey.ca/product-detail/en/T3050H-6G/497-17288-ND/7313729/?itemSeq=290474028

EDIT: Thanks for that tip on the the tab, I did see that in the datasheet. I'm a pretty careful guy just have a hard time getting parts together, usually ok after that. Also I have one of those laser temp sensors so I use that instead of fingers.

 

hi,
Comparing the BTA and T3050 the datasheet says 15mOhms for both when conducting, so they will heat to about the same body temperature.
It's just the T3060 can handle a higher current.
You could easily knock up a smallish U shaped heat sink using say 20/22 Gauge Aluminium.
or buy a TAB type ready made sink.
The only sure way is to run it for a couple of minutes and then disconnect and measure the body temperature.

E

 

I was wondering if that was the case but I will test and see. I will post results for all types. I probably won't use the BTA as it's just a Standard triac, these other 3 are Alternistor - Snubberless which I read should be better for inductive loads. The name is a little misleading, I perked up and thought hey I won't need the snubber but then they show a diagram with a snubber added lol.

 

One thing I noticed is the junction temp is 125 on BTA and 150 on others so that should help a little no?

 

Thats only 25C, if your ambient room temperature can climb by 25C , that will soon cancel that advantage out.
Those tempr curves are related to the ambient tempr.
E

 

I'm looking forward to see how it works, I got a few heat sinks as well if needed, Tuesday or Wednesday I will find out.

 

Please take care when mounting the Triac heatsink, use heatsink compound and and the correct isolating washers.

 

Yes I'll be sure nothing is shorting out. To be safe I always use an extension cord to power up something thing like this, in case it goes bang nothing can hit me.

 

Hi Eric, sorry to bother you once again, I'm going to have to setup a payroll and put you on it. I think I may have screwed up and haven't even started haha...

I've been reading some info about snubbers trying to calculate the max transient voltage it can block and came across a PDF that says the following.

Choose a resistor that’s noninductive. A good choice is a carbon composition resistor. A carbon film resistor is satisfactory unless it’s trimmed to value with a spiral abrasion pattern. Avoid wirewound because it is inductive.

Well I choose a thick film resistor, I've searched and it appears these are non-inductive so I think I'm ok there but should I be changing the type to a carbon composition or something else? I went with this type at 2/3w for all three resistors.

Part# in datasheet - ERJP08
https://industrial.panasonic.com/cdbs/www-data/pdf/RDO0000/AOA0000C331.pdf

 

hi rob,
I would not use carbon or wire wound resistors, thick film do work OK, in that application.

For ref only, you can get non inductive wire wound resistors, they are wound so that two windings cancel out any inductance, you don't need that type.

E

 

Thanks Eric!

 

I just my package of parts, connected things up and yay it works.

One thing I'm disappointed about is the heat generated, definitely more than expected. It appears you were right Eric, using a 30 amp Triac doesn't appear any cooler than a 16 amp Triac with a load of 3.4 amps. One my temp gun it gets to 95 C pretty quick, I shut it down at that point. I put on a heat sink rated at 23.00°C/W which dropped the temp down to around 60 C. After I let it run a few minutes I power it down and then touch the heat sink and it's obviously still too hot to touch. I see other heat sinks as low as 2.60°C/W Natural but that might be overkill.

I'm trying to run calculations based on heat sinks which I think I have figured out (youtube is great) but I'm unable to calculate the power dissipation (PD) which is needed to figure out the heat sink. Everything on PD is related to resistors and I've heard it's all the same but parameters used in examples are never in my datasheets. I know it's there but I'm lost.

Any ideas how I might calculate PD for a Triac?

 

hi Rob,
There is this a AAC Thread
https://forum.allaboutcircuits.com/threads/calculating-power-dispated-across-triac.114865/

also these PDF's [ light bed time reading ]
Do you have a photo image of your sink, to post.?
E

 

Thanks Eric! I will go over those shortly, first I need to get some food or I will drop. I will post some pics of my setup as well and different heat sinks.

 

hi Rob,
Check this out, may help with the maths.
https://www.daycounter.com/Calculators/Heat-Sink-Temperature-Calculator.phtml

E

 

Hi Eric, thanks for all the info, very helpful. The first paragraph knocked me out and I was almost hoping not to wake up lol... I did get through it though and ran my numbers. I used the PDF from Philips on calculating the power and then used some formulas from a guy in a youtube video to calculate the heat sink and junction temperature.

This is the Triac I plan to use.
http://www.ween-semi.com/sites/default/files/2018-10/bt139-600.pdf

To summarize I based this on a max of 7.08 amps or 850w at 120v AC.

Max power dissipation = 8.51w

Without heat sink junction temperature would get to 540.6c. Well I think something else would happen first

The smallest heat sink I could use at max junction (125c) is 6.29 c/w

I found a heat sink rated at 2.6 c/w @ Natural, using this the max junction temp would be 83.6c

I attached a PDF with all my math, some things you just have to write with pen and paper. There's some pics as well.

For anyone following along here's some links.

This is PDF Eric posted.
https://forum.allaboutcircuits.com/attachments/1760767-pdf.175720/

This is a video explaining how to calculate the heatsink and junction temperatures

This is a nice calculator for all these crazy formulas.

https://www.symbolab.com/solver/square-roots-calculator/

This is the best heat sink I found but its huge so I'm going to try and figure out another solution. I'm thinking maybe a long piece of aluminum and attach all 8 Triacs to it. This way it would be able to handle way more heat then it would ever see as all 8 Triacs can only put on 12 amps max together.

https://www.digikey.ca/product-deta...yd-corporation/530002B02500G/HS380-ND/1216384

 

hi rob,
You have been busy!
Have you tried running a single triac using this sink option from your images.?
E

 

It wasn't easy but those links you posted got me over the finish line. The heat sink you point out is what I'm currently using, the other 2 smaller ones didn't help much. But this one isn't good enough, I'm only pushing 3.7 amps through it, after 5 minutes my gun can show 90c and it takes a minute before I can even get close to touching it. So with 7 amps it'll probably melt down, I'm not even going to try it.

 

hi,
At 7A RMS, the dissipation in the Triac is approx 10Watts.
At 3.7A RMS its approx 3Watts.

Your temperature readings are very high for those wattages, with a heat sink.???
E

 

Hi Eric, I did run the numbers for 3.7a as well, I got about 4.3w which would need a 22 c/w heatsink which is probably close to what I have. I see that number is common for small TO-220 heat sinks so maybe the temps are normal. My temperature gun is on the cheap side so probably not too accurate either. I also exaggerated on the cool down time, it's more like 20 seconds till I can start to touch it. Everything is setup nicely so I can't see it putting out more heat than it should?