Triac Question

D

Dan 19 years ago

I have an conventional AC motor control circuit that uses an MOC3012 driving a BTA06 sensitive gate triac. The design has been working well for several years, however it has a random problem where the motor "bumps" or rotates slightly and randomly upon initial power-up. The motor drives a gearbox whose output shaft must not move unless commanded to do so. I've made sure the MOC3012 is not being turned on during power-up. I've also tried loading the triac's gate to gnd, A1, A2, etc. with a variety of resistors with no improvement. I swapped the MOC3012 for an MOC3043 zero-crossing driver and the problem improved considerably, but not completely. As you might have guessed, the power-up 'bumping' of the motor occurs more frequently if there is a noisy load on the same circuit as my controller. It really misbehaves with flourescent lights and soldering irons. My customer will no longer accept any movement of the gearbox shaft, however slight.

Any suggestions?

TIA, Dan

Vote

P

Phil Allison 19 years ago

"Dan"

** Which version - there are lots.

** Read the data sheet for your triac.

formatting link

Add a good snubber, use a "snubberless " version.

........ Phil

Vote

S

Spehro Pefhany 19 years ago

Swap the MOC3012 for something like a MOC3082, add a snubber if that doesn't work. And make really sure that your drive circuit isn't triggering the optoisolator.

Best regards, Spehro Pefhany

"it\'s the network..." "The Journey is the reward" speff@interlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com

Vote

D

Dan 19 years ago

Thanks for the quick replies and good advice guys.

I put in a 100 ohm/0.1uF snubber and so far so good. The glitch is random, but seems to have ceased. Spehro, I think you nailed this one! I did consider the snubber, but discounted the idea thinking the problem was gate related rather than across the anodes of the triac.

Thanks again for your help. Regards, Dan

Vote

J

john jardine 19 years ago

Sounds like you've got a single half cycle switch through. Most likely time for it to happen, is when the supply is switched on at the peak of the incoming voltage waveform. In the UK this can amount to a +/- 340V step edge turning up at the triac main terminal over the space of a few uS. The triac's dv/dt rating is massively exceeded and it switches on by itself for the rest of that single half cycle. Hence the twitching of a motor or lamp etc. This problem is completely separate from the normal action of the gate terminal and trigger pulse.

As suggested, you need a snubber across the triac. Something like 0.1uf or

0.22uF in series with maybe a 100 ohm resistor. Best to just buy the snubber item as a standard part. 1nF may be capable of doing the job but better safe than sorry. Idea is that the capacitor instantly passes the nasty voltage edge across to the other main terminal of the triac. Essentially both sides of the triac are lifted by the voltage edge at the same time, hence it cannot crowbar due to excessive dv/dt across it. For the same reason you may also need another snubber across the opto triac.

One other technique is available which is to "prewarn" the triac of the incoming waveform, so it never sees a switch-on transient. This method simply bypasses the contactor/relay with a high value resistor.

Posted via a free Usenet account from http://www.teranews.com

Vote

J

Jan Panteltje 19 years ago

On a sunny day (Sun, 15 Apr 2007 01:04:44 GMT) it happened Dan wrote in :

Fist thing you need to figure out is:

1) Is the big triac firing by itself or 2) Is the controller sending pulsens when it should not.

In case (1) make sure you also have overvoltage proctection across the triac (VDR perhaps), proper RC snubbers, the works. In case 2 you will have to scope for supply noise, pickup by wires, etc, use RF (metal) shielding perhaps.

Vote

C

cbarn24050 19 years ago

A triac is not the best device for the application, the current and voltage waveforms are out of phase so when the triac turns off at zero current there is an instantanious voltage across it which can fire it again. Hence the need for a big snubber network. There are "special" triacs made by Telcon that are supposed to address this problem. A pair of thyristors is the best way. A zero crossing MOC shouldnt be correct for the same reason. Your problem seems to be EMI. You should have a filter/surge limiter on the AC supply, a MOV accross the triac and a triac that has a voltage rating of at least twice the MOV voltage.

Vote

P

Phil Allison 19 years ago

** Groper F****IT Alert !!!

** It is perfectly OK.

** Probably not by very much.

** No serious issue.

** Yeah - called snubberless.

Stale news here - f*****ad.

** Not true at all.

** Zero crossing is OK - less volts at switch off.

** No it is not - you f****ng tenth witted prick.

The OP's main problem occurs at switch on.

** Go SFA to do with the fast voltage rise ( dv/dt ) at switch on.

** No need with an AC motor load and a simple snubber on the triac.

...... Phil

Vote

S

Spehro Pefhany 19 years ago

It's far more likely to be related to the optoisolator than the triac.

Both the breakdown voltage rating and the dv/dt rating of the optotriac in the optoisolator are far less than the WORST version of the BTA06. It's rated for only 10v/usec (less at higher temperatures than 25'C- down to less than 2V/usec at 100'C) and 250V Vdrm, whereas the MO3083 is rated for 800V Vdrm and guaranteed not to turn on with dv/dt of 600v/usec (1500v/usec typical, both at 25C).

Best regards, Spehro Pefhany

"it\'s the network..." "The Journey is the reward" speff@interlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com

Vote

D

Dan 19 years ago

The production boards are using MOC3012 drivers which are spec'd at

12V/uS typ @ 25C. Because of UL approval, swapping the '3012 for another driver such as the '3083 as you suggest is a less desirable option. Even though they are both UL approved components, the cost to update the UL paperwork for the part change is around $1,500. I know... it sounds crazy to me too.

My BT06-400C datasheet shows a dVdt of 200V/uS. I have not tried a snubber on the driver's output yet. I did however short the opto's input to guarantee that the driver is not getting turned on accidentally via the micro.

Vote

Triac Question

Join the Discussion

Didn't find your answer?