Page 1 of 2 12 LastLast
Results 1 to 10 of 15

Thread: IGBT vs Triac

  1. #1
    Join Date
    Dec 2011
    Posts
    7,092
    Post Thanks / Like

    Default IGBT vs Triac

    I am curious as to what may be the downside in using a IGBT to dim an AC load .
    From what I have seen , analogWrite can be used to pin directly and interrupts along with ZC
    can be put aside.

    Anyone have experience with IGBT and dimming that may shed some light ?

  2. #2
    Join Date
    Dec 2012
    Location
    Newtown CT
    Posts
    5,078
    Post Thanks / Like

    Default Re: IGBT vs Triac

    It is a transistor based dimmer. Since it can be turned off at any time, it can use an unsynched PWM signal to control intensity.


    2021 New Tune-to sign. New 40x27 matrix at the house. Retiring the Strip based matrix. Updating some of the old window frames to the new house. Adding two new songs.
    2020 Full sized show reworked for the new location. Only adding (famous last words) 13 RBLs that I finally got converted to using pixels
    2019 - Just moved into a new home (yet another change of plans). Will be dim but not dark. Too much to do at the new place to leave time for a show. Dim show (3000 pixels) had regular visits most nights.
    https://www.youtube.com/user/MartinMueller2003

  3. #3
    Join Date
    May 2007
    Posts
    5,571
    Post Thanks / Like

    Default Re: IGBT vs Triac

    1) It can't adequately control an AC load without rectification (like an SCR, but unlike a Triac).
    2) Assuming that not all of the logic would be operating AC-hot, it would need a different type of opto-coupler such as a 4N35 or 6N137, and therefore would need a low-voltage power source on the AC-hot side.
    3) Depending on how the IGBT is driven, there may still be a need for AC synchronization.
    4) Depending on how the IGBT is driven (i.e. low frequency or high frequency) it may need filter logic to minimize the sending of high-frequency noise back into the AC line (not that thyristor-type SSRs don't have their own issues here).

    IGBT's are not a panacea, and may not be cheaper when all the components added around it are taken into cnsideration.
    Phil

  4. #4
    Join Date
    Dec 2011
    Posts
    7,092
    Post Thanks / Like

    Default Re: IGBT vs Triac

    Quote Originally Posted by P. Short View Post
    1) It can't adequately control an AC load without rectification (like an SCR, but unlike a Triac).
    2) Assuming that not all of the logic would be operating AC-hot, it would need a different type of opto-coupler such as a 4N35 or 6N137, and therefore would need a low-voltage power source on the AC-hot side.
    3) Depending on how the IGBT is driven, there may still be a need for AC synchronization.
    4) Depending on how the IGBT is driven (i.e. low frequency or high frequency) it may need filter logic to minimize the sending of high-frequency noise back into the AC line (not that thyristor-type SSRs don't have their own issues here).

    IGBT's are not a panacea, and may not be cheaper when all the components added around it are taken into cosideration.
    Thank you for this clear explanation .
    The triac method for dimming this 200W flood does not appear to be a sound solution . Requiring a parallel load more than a 5w incan in order to get stable low end dimming is a no go.

    I was contemplating trying a IGBT and opto-couple like this fellow has done , but the hot line @ 4:15 has me concerned .


  5. #5
    Join Date
    May 2007
    Posts
    5,571
    Post Thanks / Like

    Default Re: IGBT vs Triac

    The thing that gives me pause is that we don't really know what is causing the difficulties that you experience. It's my guess is that the characteristics of the PM2014 controlling the SCRs are not well understood, and it's my conjecture that something about that part or the circuit around it is what's causing troubles. Also, the light that is being controlled in the video is an incandescent bulb, which does not behave in the same manner as LED lights, let alone the floods that you're working with.

    So, if you want, go ahead and try the IGBT circuit, ensuring that you're using opto-coupler as shown later in the video. Also, the component values should probably be a bit different from those in the video for 110VAC/60Hz operation. It'll be a shot in the dark, but what the heck.
    Phil

  6. #6
    Join Date
    Dec 2011
    Posts
    7,092
    Post Thanks / Like

    Default Re: IGBT vs Triac

    Quote Originally Posted by P. Short View Post
    The thing that gives me pause is that we don't really know what is causing the difficulties that you experience. It's my guess is that the characteristics of the PM2014 controlling the SCRs are not well understood, and it's my conjecture that something about that part or the circuit around it is what's causing troubles. Also, the light that is being controlled in the video is an incandescent bulb, which does not behave in the same manner as LED lights, let alone the floods that you're working with.

    So, if you want, go ahead and try the IGBT circuit, ensuring that you're using opto-coupler as shown later in the video. Also, the component values should probably be a bit different from those in the video for 110VAC/60Hz operation. It'll be a shot in the dark, but what the heck.
    I like your style there

    I am actually testing with a string of standard AC leds , a string of incandescent minis , a single 5w christmas retro incan bulb and the 200W flood .
    The #1 thing I've learned so far is that using the HW.timer code with the esp is so far superior to the software in how leds respond with this AC 1 channel module.
    There is more learned in failure than success !
    Thanks for the boost . Going to round up the parts an see what results.

  7. #7
    Join Date
    Dec 2011
    Location
    UK S80 postcode
    Posts
    1,535
    Post Thanks / Like

    Default Re: IGBT vs Triac

    Hi angus, before you spend to much time on alterative components have a look back at this table in some code you posted.

    http://doityourselfchristmas.com/for...341#post539341

    I haven’t looked into it but it looks like #ifndef RBDMCUESP8266_H is a dimming curve.

    I can only guess that you had to set the lowest value to -1 to switch off the light as this table has no 0.

    It might be worth changing these 100 values to:
    75, 75, 74, 74, -> 26, 26, 25, 25

    This would be more appropriate for the circuit of this LED unit but won’t have an off.

    Then through trial and error you can work out the lowest value for ‘25’ as above that is a stable lowest on and the highest value for ‘75’ that is a stable on and adjust the table accordingly

    Whilst I nearly always make comments / recommendations on the forum, in this case I strongly suggest that this is a software issue and without looking into this, you may dive down a rabbit hole that would be interesting to follow experiencing new components and bringing them to the attention of forum members. I believe it would involving you expending more time and money than is necessary and I’m not convinced would return the results you want.

  8. #8
    Join Date
    Dec 2011
    Posts
    7,092
    Post Thanks / Like

    Default Re: IGBT vs Triac

    Quote Originally Posted by Barnabybear View Post
    Hi angus, before you spend to much time on alterative components have a look back at this table in some code you posted.

    http://doityourselfchristmas.com/for...341#post539341

    I havenít looked into it but it looks like #ifndef RBDMCUESP8266_H is a dimming curve.

    I can only guess that you had to set the lowest value to -1 to switch off the light as this table has no 0.

    It might be worth changing these 100 values to:
    75, 75, 74, 74, -> 26, 26, 25, 25

    This would be more appropriate for the circuit of this LED unit but wonít have an off.

    Then through trial and error you can work out the lowest value for Ď25í as above that is a stable lowest on and the highest value for Ď75í that is a stable on and adjust the table accordingly

    Whilst I nearly always make comments / recommendations on the forum, in this case I strongly suggest that this is a software issue and without looking into this, you may dive down a rabbit hole that would be interesting to follow experiencing new components and bringing them to the attention of forum members. I believe it would involving you expending more time and money than is necessary and Iím not convinced would return the results you want.
    I gave this a go as you suggested and it seems a bit more stable .

    I just remarked the original to fall back on.
    Definitely less flicker than just tweaking the web server Range slider value .
    I am not sure what's required of the buffer table to keep it happy without breaking it ,so I just filled in the top.
    I have stable light output now on the bottom end with ramping start at 24.

    Thank you for this tip to the puzzle.

    Code:
    /*static const uint8_t powerBuf[] = {
        100, 99, 98, 97, 96, 95, 94, 93, 92, 91,
         90, 89, 88, 87, 86, 85, 84, 83, 82, 81,
         80, 79, 78, 77, 76, 75, 74, 73, 72, 71,
         70, 69, 68, 67, 66, 65, 64, 63, 62, 61,
         60, 59, 58, 57, 56, 55, 54, 53, 52, 51,
         50, 49, 48, 47, 46, 45, 44, 43, 42, 41,
         40, 39, 38, 37, 36, 35, 34, 33, 32, 31,
         30, 29, 28, 27, 26, 25, 24, 23, 22, 21,
         20, 19, 18, 17, 16, 15, 14, 13, 12, 11,
         10,  9,  8,  7,  6,  5,  4,  3,  2,  1 */
    
    static const uint8_t powerBuf[] = {
        75, 75, 75, 75, 75, 75, 75, 75, 75, 75,
         75, 75, 75, 75, 75, 75, 75, 75, 75, 75,
         75, 75, 75, 75, 75, 75, 74, 73, 72, 71,
         70, 69, 68, 67, 66, 65, 64, 63, 62, 61,
         60, 59, 58, 57, 56, 55, 54, 53, 52, 51,
         50, 49, 48, 47, 46, 45, 44, 43, 42, 41,
         40, 39, 38, 37, 36, 35, 34, 33, 32, 31,
         30, 29, 28, 27, 26, 25, 24, 23, 22, 21,
         20, 19, 18, 17, 16, 15, 14, 13, 12, 11,
         10,  9,  8,  7,  6,  5,  4,  3,  2,  1
    };

  9. #9
    Join Date
    Dec 2011
    Posts
    7,092
    Post Thanks / Like

    Default Re: IGBT vs Triac

    Well I have gone back to the hw.timer sketch and with the parallel load the dimming is exactly what should be expected .
    The range is perfect from off -> to full on just as if dimming dumb rgb , incans or pixels. The only shortfall is the decay time lags slightly.

    The AC dimmer lib from the manufacture is broken and without edits is really useless firmware .
    All in all it was a good learning experience as I searched for a better method .and found it with hw .timer
    I just need to figure out what to use as a parallel load that is not light emitting .

    In the meanwhile I am going to pursue the parts for IGBT and see what shakes out .

  10. #10
    Join Date
    May 2007
    Posts
    5,571
    Post Thanks / Like

    Default Re: IGBT vs Triac

    Do you know if you really need a full 5W load (if that was your latest choice)? A 3KΩ, 10W resistor would dissipate 5W across a 110VAC line with about 100% margin as far as power dissipation goes, although it would get hot. If you don't really need 5W and 1W would do, then a 15KΩ, 2W resistor would work. It would not surprise me if the 15KΩ resistor would do what you need.
    Phil

Page 1 of 2 12 LastLast

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •