TC4 with aArtisan... | [411] |
Transparent Fluid... | [125] |
TC4 Roasting App | [69] |
Modifying DASH 14... | [35] |
RobotDyn AC Light... | [34] |
PWM AC Light Dimmer Module 50Hz 60Hz
|
|
greencardigan |
Posted on 02/19/2019 6:44 PM
|
![]() 1 1/2 Pounder ![]() Posts: 1133 Joined: November 21, 2010 |
Hi, I purchased one of these 'PWM AC Light Dimmer Module 50Hz 60Hz' boards from Tindie for testing with the TC4 PWM output on IO3. https://www.tindi...50hz-60hz/ Has anyone used these successfully? I'm having three issues that I'm struggling with if anyone has any suggestions. 1. The output pulses and stutters. This occurs both on my TC4 and Raspberry Pi and at various PWM frequencies. 2. The output does not seem to be linear in terms of the power output. Larger power output changes in the middle portion of the PWM duty cycle range. 3. The power output sometimes switches on for a fraction of a second when the AC power is switched on initially. Even when the PWM duty cycle is set to 0. |
|
|
btreichel |
Posted on 02/19/2019 10:27 PM
|
![]() 1/4 Pounder ![]() Posts: 161 Joined: May 07, 2007 |
is it zero crossing or random fire? And yes, the power won't be linear. The area under a sine wave isn't linear, that area is power. So power won't be linear; unless you linearize it. There is no guarantee on the state of the output when you 'boot; the micro. its 50 or 60 hz labeled because thats suppose to be line frequency. |
|
|
renatoa |
Posted on 02/20/2019 1:17 AM
|
![]() Administrator ![]() Posts: 1626 Joined: September 30, 2016 |
Used it, less or... less successful ![]() How you connected this board to IO3 ? If directly, then won't work. It is actually misleading advertised as PWM dimmer, because is rather a voltage controlled dimmer. Also, the control logic is reversed, power raise when control voltage decrease. And yes, the output is not linear, but acceptable... More exactly, the zero output is somewhere in the 2.85V control voltage (CV) ballpark, and the power increase lowering this voltage down to about 1V, where is full 100%. I explained here some basics of the functionality: https://forum.hom...post_67821 First, it needs a prepolarisation of CV+ input, that led to OC1 pin 5, with a pot, to establish a 2.85V voltage, for zero output heater. Next, you have to inject your PWM a bit different than expected, not as positive pulses, as I guess you have now coming from IO3, but shorting the CV+ input (pot cursor) to the ground. So you need to change IO3 pinMode to open collector GPIO (OUTPUT_OPEN_DRAIN), not OUTPUT as is now. (PWM_IO3::Setup) If you want keep PWM16 source unchanged you have to use a transistor, almost same circuitry as for SSR outputs: IO3 into transistor base with a resistor, and collector to CV+ Alternative to all above fuss, as I explained in the link, you can drop this hardware simulation of PAC control, and use the ZCD and SSR blocks from this dimmer with TC4 for PAC or ICC. These blocks schematic is here: https://esp8266-p...ic-0_0.jpg ...I can help identify interest points on PCB, if needed. |
|
|
greencardigan |
Posted on 02/20/2019 5:33 AM
|
![]() 1 1/2 Pounder ![]() Posts: 1133 Joined: November 21, 2010 |
Thanks but are you referring to this other board? https://www.tindi...r-mpdmv41/ I was hoping one of these boards would be a simple replacement for the usual ZCD and random fire SSR |
|
|
renatoa |
Posted on 02/20/2019 6:26 AM
|
![]() Administrator ![]() Posts: 1626 Joined: September 30, 2016 |
oooups... Yes, my rant was been about a different board that you asked in OP. As I see, the OP board is also similar, a ZCD block, a SSR block, both can be clearly outlined, and an ATTINY processor that is expected to run a PAC code. How well is this done... this is the question... |
|
|
greencardigan |
Posted on 02/20/2019 2:21 PM
|
![]() 1 1/2 Pounder ![]() Posts: 1133 Joined: November 21, 2010 |
No problems. I will continue testing it and see how I go. There is also another board that I might try. https://www.tindi...ler-board/ |
|
|
renatoa |
Posted on 02/20/2019 3:09 PM
|
![]() Administrator ![]() Posts: 1626 Joined: September 30, 2016 |
Issues with PAC, or what? seems you want to make Arduino processor life easier ... What's wrong with slow PWM, relay style ? If the power levels range is >50% duty, then using the diode trick it can be made even smother than a simple on-off. Edited by renatoa on 02/20/2019 3:14 PM |
|
|
greencardigan |
Posted on 02/20/2019 4:19 PM
|
![]() 1 1/2 Pounder ![]() Posts: 1133 Joined: November 21, 2010 |
I have no real issue with the current PAC setup with ZCD and random fire SSR. But it would be cheaper to replace with a PWM AC Dimmer. Also, I am testing a setup with a Raspberry Pi running Artisan but without a TC4. Using cheap MAX31855 SPI thermocouple boards for temperature input and PWM out using the Raspberry Pi header pins. So slow PWM will be fine for a heater via zero crossing SSR, but AC motor control still needs phase angle control. |
|
|
renatoa |
Posted on 02/20/2019 11:53 PM
|
![]() Administrator ![]() Posts: 1626 Joined: September 30, 2016 |
Cheaper than $3.49 ? https://robotdyn....-110v.html Don't worry about the 5A limit, it can be increased, as I wrote in another post. Forget about MAXes, the resolution and sample time are pathetic. You can't compute a decent RoR with 0.25 degree res. You need more than 30 seconds averaging window for 10% precision. The pre-amp'ed MCP3424 solution used for ESP board give phidget class results for 2 channels for the price of a single MAX. |
|
|
mg512 |
Posted on 03/05/2019 11:55 AM
|
![]() 1/4 Pounder ![]() Posts: 173 Joined: March 04, 2018 |
Quote greencardigan wrote: 1. The output pulses and stutters. This occurs both on my TC4 and Raspberry Pi and at various PWM frequencies. 2. The output does not seem to be linear in terms of the power output. Larger power output changes in the middle portion of the PWM duty cycle range. 3. The power output sometimes switches on for a fraction of a second when the AC power is switched on initially. Even when the PWM duty cycle is set to 0. I haven't used the PWM dimmer board for my roaster (since I don't use an AC fan), but I've had one for a while to play around with. So far I've used it mostly with the pump in my espresso machine. But I've also just hooked it up to a hot air gun to check that it works with a fan motor as well, just in case there's any difference between a pump and a motor. I haven't encountered either 1 or 3. What exactly do you mean by "pulses and stutters"? As in, the fan speed noticeable changes? Or you're looking at the output waveform with an oscilloscope, and that's looking wrong? I haven't done the latter, but fan/pump speed definitely seems constant for me. Maybe you have a bad board? Have you tried contacting the Tindie seller? They've been very responsive for me in the past. Or is there some difference between different kinds of motors that I'm not considering, that could be causing it? As for 2, I think that's to be expected, given that you're just chopping off parts of the sine curve (which aren't all equal). Does the aArtisan sketch compensate for this when used in PAC configuration? If so, that could be easily enabled for PWM configuration as well I imagine. Quote Don't worry about the 5A limit, it can be increased, as I wrote in another post. Indeed. I've confirmed this with the seller as well, all that's needed is a bigger heatsink on the TRIAC. |
|
|
renatoa |
Posted on 03/05/2019 12:27 PM
|
![]() Administrator ![]() Posts: 1626 Joined: September 30, 2016 |
Yes, aArtisan uses a phase_delay mapping table, to ensure the linearity of power output versus control input. Other words, when input command is, say 20 vs 40%, the chopped pulses length is not double, as 2 vs 4 ms (for 50 Hz), but adjusted using that mapping table to have the area under chopped sine double, thus duty cycle (output power) be proportional with input. |
|
|
greencardigan |
Posted on 03/05/2019 6:58 PM
|
![]() 1 1/2 Pounder ![]() Posts: 1133 Joined: November 21, 2010 |
This video is with a 600W vacuum motor but I get the same issue with an incandescent light. No I haven't contacted the seller. I just assumed it was a design issue. I've already purchased an alternative board from Tindie so will test it out too when it arrives. Edited by greencardigan on 03/05/2019 7:14 PM |
|
|
mg512 |
Posted on 03/06/2019 3:23 AM
|
![]() 1/4 Pounder ![]() Posts: 173 Joined: March 04, 2018 |
Quote greencardigan wrote: This video is with a 600W vacuum motor but I get the same issue with an incandescent light. No I haven't contacted the seller. I just assumed it was a design issue. I've already purchased an alternative board from Tindie so will test it out too when it arrives. Ah. No, the one I got doesn't do anything like that. |
|
|
greencardigan |
Posted on 03/06/2019 4:33 AM
|
![]() 1 1/2 Pounder ![]() Posts: 1133 Joined: November 21, 2010 |
Ok I may contact the seller then. Are you on 50Hz or 60Hz? |
|
|
mg512 |
Posted on 03/06/2019 5:05 AM
|
![]() 1/4 Pounder ![]() Posts: 173 Joined: March 04, 2018 |
50Hz, 230V. |
|
|
renatoa |
Posted on 03/06/2019 5:24 AM
|
![]() Administrator ![]() Posts: 1626 Joined: September 30, 2016 |
The 50/60 autodetection is tricky, with some networks being extremely noisy... I have here circuits where I can get even 10 false zero cross triggering in the -15+15V interval. This parameter should be hard coded in firmware, or switchable on board with a jumper. |
|
Jump to Forum: |
Thread | Forum | Replies | Last Post |
---|---|---|---|
RobotDyn AC Light Dimmer Module | Dataloggers/Controllers/Rate of Rise Meters | 35 | 03/05/2021 10:55 PM |
RobotDyn Transistor DC Relay Module | Dataloggers/Controllers/Rate of Rise Meters | 2 | 01/14/2021 3:41 AM |
Hottop 2k+ Light Roast Profile | HotTop Roaster | 9 | 09/21/2020 3:07 PM |
90% of my roasts, even light City roasts, have an underlying burnt/acrid taste. What gives?! | Roasting Coffee | 27 | 08/04/2020 6:50 AM |
Hottop (Light) Roast Profile, Baked Coffee | HotTop Roaster | 3 | 01/12/2020 2:19 AM |