Post by thread:[OT]: TRIACs in parallel

I have an application wherein the PIC is to drive an AC load at 240 volts, about 10A. For certain reasons, I cannot use a relay and have to use a TRIAC instead. Instead of using a TRIAC with a high current capacity of the required value, can I 'parallel' two or three of them safely, and expect them to share the load current? I have tried this technique briefly, and nothing blew, but how safe / reliable is it? I would appreciate comments on this. Regards, Anand -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

At 09:44 PM 6/16/01 +0530, you wrote: >I have an application wherein the PIC is to drive an AC load at 240 volts, about 10A. For certain reasons, I cannot use a relay and have to use a TRIAC instead. Instead of using a TRIAC with a high current capacity of the required value, can I 'parallel' two or three of them safely, and expect them to share the load current? I have tried this technique briefly, and nothing blew, but how safe / reliable is it? It isn't, not very. Devices like triacs tend to "hog" current when there is a temperature differential. You *could* do it, by mounting them all on a common heat sink and arranging some kind of ballast resistors, but why bother? For a commercial application with a (real) 10A (resistive) load, I'd use an inexpensive 25A triac, most likely, or even a 40A unit. Even then, protection (fusing) can be an issue. You can cheap out and just use the fuse to protect wiring and assume the triac will go if there is a short. You generally need a *lot* more margin than on relay contacts for long-term reliable operation in a production situation. Keep them cool, and keep in mind that failure of the eutectic die bond (leading to overheating and failure "on") is one of the common failure modes, so thermal cycling should be minimized. You'll have 10W, give or take, to get rid of. Best regards, =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Spehro Pefhany --"it's the network..." "The Journey is the reward" spam_OUTspeffTakeThisOuTinterlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com Contributions invited->The AVR-gcc FAQ is at: http://www.bluecollarlinux.com =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Yeah, margin on relay contacts - I wish I had any. We often use 10 amp relays on 13 amp loads! We can get away with this because the loads are resistive, and the relays are rated for inductive loads. I've actually approached manufacturers and had them get a special UL listing for resistive loads only, so we can squeeze every last watt out of the relay contacts. Needless to say, this isn't on a rocket science, nuclear, medical, downhole or other mission critical applicaiton, just a toaster. I'd agree, don't parallel Triacs, it's asking for trouble. One gets hot, it'll hog all the current and blow. I stay away from Triacs as much as possible, they are nothing but trouble. Relays, which everybody thought were history 20 years ago, are doing more volume than ever. They run cool, don't take up as much room as a heatsink, and last for millions of operations ( I life tested them) even when overloaded. -- Lawrence Lile {Original Message removed}

It is generally bad form to parallel semiconductor components for larger loads than a single component can handle without long and hard consideration to what exactly is going to happen. This is where the rubber (theory) hits the road (real life). In theory each triac will handle exactly one-third of the current. This is due to the theory that each triac has exactly the same on resistance as the other traics. In practice no two components are exactly alike, and so one component will handle more current than the other two. This additional current will cause additional heat, lowering the on resistance of the affected triac, and forcing it to handle even more current. While this works fine on the bench, it will eventually lead to that component's eventual heat failure, which will then force (suddenly) a larger load on the other two (assuming the first failed open), and they too will fail. The last one to go will probably let you know quite audibly. To avoid this, mount them all on the same heat sink, and put an additional resistor in series with each triac with its resistance being a magnitude greater than the individual on resistance of the triacs. Given that the extra expended effort and cost of additional components (not to mention the associated headaches and problems) is far greater than specifying the proper part in the first place, there really is no reason to try to make three components do the work of one. It appears, though, that you are restricted to making his design a certian way, and you've doubtlessly considered using a properly rated part. What restriction is keeping you from using a 15 or 20A TRIAC? Why can't you use a relay? -Adam Anand Dhuru wrote: {Quote hidden}>I have an application wherein the PIC is to drive an AC load at 240 volts, about 10A. For certain reasons, I cannot use a relay and have to use a TRIAC instead. Instead of using a TRIAC with a high current capacity of the required value, can I 'parallel' two or three of them safely, and expect them to share the load current? I have tried this technique briefly, and nothing blew, but how safe / reliable is it? > >I would appreciate comments on this. > >Regards, > >Anand > >-- >http://www.piclist.com hint: PICList Posts must start with ONE topic: >[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads > > > > > -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email .....listservKILLspam@spam@mitvma.mit.edu with SET PICList DIGEST in the body