Also, make sure you're not triggering the triac prematurely, i.e. at the end of the previous half cycle. That would cause it to conduct, then immediately quench.
Just out of curiosity, does your blower's switch go from off to high-speed to low-speed? If so it might have an AC induction motor, rather than a universal motor.
Ah, so you actually had an SSR with the zero-cross switching in it, no wonder. Let me know if you need a part number for a suitable SSR TRIAC, that is a regular triac with the opto built in as they are quite handy and also are just like a big transistor that look like this one - but not that one, I will have to see who has stock and what. Actually the ones I have used that are very cheap are Sharp S20202s but I am having trouble tracking them down. Otherwise just just an opto-triac driving a regular triac.
Just out of curiosity, does your blower's switch go from off to high-speed to low-speed? If so it might have an AC induction motor, rather than a universal motor.
The motor definitely has brushes. As I mentioned earlier (not that I expect all of you to read all that I write), I was able to run the motor with a DC bench supply. It worked okay down to 7V. It would spin at lower voltages but it required 7V to start turning from a stopped position.
I should have taken some photos of the motor while I had the blower taken apart.
Let me know if you need a part number for a suitable SSR TRIAC, that is a regular triac with the opto built in as they are quite handy and also are just like a big transistor that look like this one - but not that one, I will have to see who has stock and what.
I'd love a link to a suitable SSR TRIAC. I purchase stuff from DigiKey and Mouser relatively often and I'd like to add a couple SSR TRIAC to my next order.
As I mentioned earlier, looking at DigiKey's search filter for TRIACs was more than a bit intimidating.
I'd love a link to a suitable SSR TRIAC. I purchase stuff from DigiKey and Mouser relatively often and I'd like to add a couple SSR TRIAC to my next order.
As I mentioned earlier, looking at DigiKey's search filter for TRIACs was more than a bit intimidating.
kwinn,
Am I missing something? The existing SSR already has zero-crossing. That's what's interfering.
You're right, that 480V SSR does have zero crossing. Most of them do, that's why I initially suggested full/half cycle control. With an induction motor it has to be full cycles on or off to avoid any DC bias, with a universal (brushed) motor half cycles or full cycles on or off will work.
At one point I thought that the SSR had been dropped and a non-zero crossing triac had replaced it so phase control could be used. A case of too many chefs methinks.
At one point I thought that the SSR had been dropped and a non-zero crossing triac had replaced it so phase control could be used. A case of too many chefs methinks.
Hey, I really like hearing all these suggestions!
Thanks a bunch for all the part suggestions. They give a good place to start looking for these types of parts.
Just found this conversation...Duane, I actually have a good bit of experience with controlling large (>500W) loads with a TRIAC (MAC12M), TRIAC-output optocoupler (TLP525G), and a couple PC817 optocouplers thrown in for feedback. I was actually controlling a 1/2hp brushed DC motor (through a bridge rectifier / 4,700µF capacitor), and it worked quite well. (For the record, I was using a PIC16F747 coded in ASM.)
Personally, I would recommend a slightly larger TRIAC than "minimal", as with chopped-wave, you will have a much higher "surge" current when the TRIAC cuts in midway through the wave. I did a quick parametric search on Mouser, and their catalog #771-BT138-600 will be more than adequate for almost anything you throw at it. 12A / 600V. Alternatively, there's #771-BT137-600 which only costs 70¢ apiece. Note that a TRIAC has a forward voltage drop of 1.3-1.65v, so at 10A, that's 13W of heat to dissipate. It'll likely require a heatsink of some sort. And (cue school of hard knocks)...that heatsink will be electrically "hot", unless you have some isolated heatsink mounts. Don't ask me how I know ;-)
If you don't have any optotriacs in your "parts bin", you can get one (WITHOUT a zero cross-detect circuit) from Mouser for 32¢: #859-MOC3022M. This CANNOT have a zero-cross circuit in it, or you'll be limited to two speeds: on and off. A TRIAC when triggered, remains "on" until the current flowing through it drops below the holding current (5-20mA on the BT138 referenced above). A zero-cross circuit in the optotriac will prevent you from triggering the TRIAC anywhere but a zero crossing. Perfect for on-off control by a CPU that isn't watching the AC wave...and a complete wreck for a CPU trying to chop the AC line.
As far as the zero-cross detect circuit, I won't touch that, as you've already assembled it. (It should work just fine.) Personally, I'd have just used a much larger resistor on both sides (AC side and CPU side). As far as triggering, just go with a calculated timebase. You'll need to trigger your timer on both the rising and falling edges of the signal, or you'll have a VERY noisy motor. You'll probably want to hardcode limits on both ends; if you "run off the end", it'll have some very interesting results.
Personally, one of my most memorably nerve-wracking moments was debugging my TRIAC motor controller. Inching power up by pressing keys on my computer keyboard...and then having it pretty much explode in my face (blew out a power Darlington transistor, two optocouplers, and the TRIAC) when I forgot that a standard transistor-output optocoupler has a breakdown voltage of about 50v...and I was going for 90v. Not to mention the 4,700µF, 100v capacitor that all-too-frequently got blasted up to 180v for a few seconds when I'd have a code malfunction! Exciting days, alright.
Comments
Just out of curiosity, does your blower's switch go from off to high-speed to low-speed? If so it might have an AC induction motor, rather than a universal motor.
-Phil
The motor definitely has brushes. As I mentioned earlier (not that I expect all of you to read all that I write), I was able to run the motor with a DC bench supply. It worked okay down to 7V. It would spin at lower voltages but it required 7V to start turning from a stopped position.
I should have taken some photos of the motor while I had the blower taken apart.
I'd love a link to a suitable SSR TRIAC. I purchase stuff from DigiKey and Mouser relatively often and I'd like to add a couple SSR TRIAC to my next order.
As I mentioned earlier, looking at DigiKey's search filter for TRIACs was more than a bit intimidating.
I'm working on some half cycle control code.
http://www.digikey.ca/product-detail/en/lite-on-inc/MOC3063/160-1722-5-ND/670014
http://www.digikey.ca/product-detail/en/stmicroelectronics/Z0405NF0AA2/497-16470-5-ND/6003572
http://www.digikey.ca/product-detail/en/nxp-semiconductors/BT137-600,127/568-3656-5-ND/1154768
All three are under $1.00 CAD at digikey Canada.
https://www.fairchildsemi.com/application-notes/AN/AN-3004.pdf
Am I missing something? The existing SSR already has zero-crossing. That's what's interfering.
Ah ha, decided to look under opto-isolators with triac outputs and found this one in stock.
I mentioned in an earlier post that you could just use a diode to get a simple half wave for a fixed low speed. One component speed control.
You're right, that 480V SSR does have zero crossing. Most of them do, that's why I initially suggested full/half cycle control. With an induction motor it has to be full cycles on or off to avoid any DC bias, with a universal (brushed) motor half cycles or full cycles on or off will work.
At one point I thought that the SSR had been dropped and a non-zero crossing triac had replaced it so phase control could be used. A case of too many chefs methinks.
The blower actually uses a diode for its low speed mode.
Thanks for the part suggestion.
Hey, I really like hearing all these suggestions!
Thanks a bunch for all the part suggestions. They give a good place to start looking for these types of parts.
Personally, I would recommend a slightly larger TRIAC than "minimal", as with chopped-wave, you will have a much higher "surge" current when the TRIAC cuts in midway through the wave. I did a quick parametric search on Mouser, and their catalog #771-BT138-600 will be more than adequate for almost anything you throw at it. 12A / 600V. Alternatively, there's #771-BT137-600 which only costs 70¢ apiece. Note that a TRIAC has a forward voltage drop of 1.3-1.65v, so at 10A, that's 13W of heat to dissipate. It'll likely require a heatsink of some sort. And (cue school of hard knocks)...that heatsink will be electrically "hot", unless you have some isolated heatsink mounts. Don't ask me how I know ;-)
If you don't have any optotriacs in your "parts bin", you can get one (WITHOUT a zero cross-detect circuit) from Mouser for 32¢: #859-MOC3022M. This CANNOT have a zero-cross circuit in it, or you'll be limited to two speeds: on and off. A TRIAC when triggered, remains "on" until the current flowing through it drops below the holding current (5-20mA on the BT138 referenced above). A zero-cross circuit in the optotriac will prevent you from triggering the TRIAC anywhere but a zero crossing. Perfect for on-off control by a CPU that isn't watching the AC wave...and a complete wreck for a CPU trying to chop the AC line.
As far as the zero-cross detect circuit, I won't touch that, as you've already assembled it. (It should work just fine.) Personally, I'd have just used a much larger resistor on both sides (AC side and CPU side). As far as triggering, just go with a calculated timebase. You'll need to trigger your timer on both the rising and falling edges of the signal, or you'll have a VERY noisy motor. You'll probably want to hardcode limits on both ends; if you "run off the end", it'll have some very interesting results.
Personally, one of my most memorably nerve-wracking moments was debugging my TRIAC motor controller. Inching power up by pressing keys on my computer keyboard...and then having it pretty much explode in my face (blew out a power Darlington transistor, two optocouplers, and the TRIAC) when I forgot that a standard transistor-output optocoupler has a breakdown voltage of about 50v...and I was going for 90v. Not to mention the 4,700µF, 100v capacitor that all-too-frequently got blasted up to 180v for a few seconds when I'd have a code malfunction! Exciting days, alright.