topbanner.gif
Login
Username

Password




Not a member yet?
Click here to register.

Forgotten your password?
Request a new one here.
Shoutbox
You must login to post a message.

renatoa
08/10/2022 1:56 AM
nguyencoffeesupply
and RoyB, Welcome

renatoa
08/06/2022 1:31 PM
Welcome, oak202, CoffeeNutZ and Mlcharlestonsc

Strangeworth
08/04/2022 9:34 PM
Thank you!

renatoa
08/03/2022 2:09 AM
Strangeworth and BigPalm, welcome to forum

Tamesino
08/02/2022 12:01 PM
Thank you guys and gals !! Exited to be here greenman

In Memory Of Ginny
Donations

Latest Donations
dmccallum - 10.00
JackH - 25.00
snwcmpr - 10.00
Anonymous - 2.00
Anonymous - 5.00
Users Online
Guests Online: 5

Members Online: 0

Total Members: 7,627
Newest Member: nguyencoffeesupply

View Thread

Who is here? 1 guest(s)
 Print Thread
RobotDyn AC Light Dimmer Module
dpecsok
fyi, i ordered the updated version with the snubber capacitors on the board but instead received an update to the updated version ha! looks like they removed the capacitors again .... size, heatsink, and AC connections are larger; about the size of an SSR now.

trying to add a picture as an attachment but its not working...
 
iPa
As soon I switched from ZCD module bought with the TC4 shield ("ZCDboard") to the ZCD from the Robodyn module ("ZCDrobodyn") I noticed a weird behavior change with the PID control. I made many tests to confirm this difference. For now I don't have any explanation, yes I should get out the oscilloscope...
Any ideas of explanation about this difference, welcome Grin

ZCDboard result:
A bit below the curve at the beginning but reached it soon, maybe need to increase ki
i84.servimg.com/u/f84/20/23/21/77/captur27.png

ZCDrobodyn result:
A lot more oscillations in this case
i84.servimg.com/u/f84/20/23/21/77/captur28.png
 
renatoa
Is the LED on Arduino board lit both cases ? i.e. zcd working, a/c detected.

Looks like you have on-off switching in the second scenario...
 
iPa
The oscilloscope revealed some interesting things:
- The signal polarity is not the same between this 2 boards !!
- If the board is to not connected to VCC the signal is less good drived (so better to connect vcc).

Need to have a look in the code to understand the impact of this reversed polarity, and what is expected.
Maybe @renatoa have you some information about that ?


ZCDboard with vcc
i84.servimg.com/u/f84/20/23/21/77/img_8013.jpg

ZCDboard without vcc
i84.servimg.com/u/f84/20/23/21/77/img_8012.jpg

ZCDrobodyn note the inverted polarity compared to the ZCDboard (with vcc)
i84.servimg.com/u/f84/20/23/21/77/img_8010.jpg

ZCDrobodyn without vcc
i84.servimg.com/u/f84/20/23/21/77/img_8011.jpg

if someone has schematics about this 2 boards I am interested
edit1: I found schematic for the Robotdyn but not for the NZDboard (mine is a verson 6.2)

edit2: Found in the code:
Expected> logic low indicates zero cross
Interrup on falling edge> attachInterrupt( EXT_INT, ISR_ZCD, FALLING );
That means, ZCD interrupt will occurs about 1mS later with ZCDRobotdyn (1/2 period = 10mS = 10%), is that enough to explain the difference behavior, probably. Asap I will test with interrupt on the rising edge with the ZCDRobotdyn.
Edited by iPa on 03/27/2022 10:44 PM
 
renatoa
Could be related to similar processes, as investigated some posts above, starting with #51.

If not intriguing enough, in my TC4ESP fork I changed also the polarity, but for other reasons, without scoping the ZCD signal... Grin

My posts about with/without Vcc pin connected are related to robotdyn board only, never used another make, so can't comment about their behavior, without schematic.
Edited by renatoa on 03/28/2022 1:38 AM
 
iPa
Confirmed!
To use Zero Cross Detection from the Robotdyn AC Dimmer, need to trig on the rising edge. Now my TC4 PID is working as expected (and the Fan drived by the Robotdyn dimmer works very well too).

Note, both board are correctly labeled, ZCDboard (the one bought with the TC4 shield) the output is labeled “/ZC” and the output on the Robotdyn is labeled “ZC”

It should be a good idea to add a define in the code source for to use ZC from the Robotdyn (all boards with positive pulse). Should open a request on github.

How to:
edit phase_ctrl.cpp and replace (line 169) :
attachInterrupt( EXT_INT, ISR_ZCD, FALLING );
by
attachInterrupt( EXT_INT, ISR_ZCD, RISING );
Edited by iPa on 03/29/2022 5:19 AM
 
jake415
Any idea why the robotdyn might modulate power differently depending on the load?
I have the usual setup with robotdyn dimmer, TC4+, and Artisan and when I change use the slider controls on artisan while measuring the current at various levels these are the graphs. (They assume that the artisan sliders relate directly to PAC %)

I'm wondering why the fan current graph is linear while the heater graph is not, and maxes at 50%.
Any ideas? Is it just because the loads are different? Before testing this, I used the heating element at 50% and it burnt out.

For reference this is the fan
https://www.amazo...&psc=1

and heating element
https://www.maste...udes-mica/

Also, as an aside question, what PAC percentage is generally used to control the heater? Seems like it must be pretty low based on this graph.

Thanks
jake415 attached the following images:
heating_element_1.png fan.png

Edited by renatoa on 04/26/2022 3:02 AM
 
renatoa
Maybe you have a MAX_OT1 setting of 50, to limit the heater power to 50%
Because you later stated that...

Quote

... I used the heating element at 50% and it burnt out.

Please explain this... was without airflow, or why burnt...

There is no recommended power percentage, depends on roasting phase and machine build.
For FB and radiant machine the power is pretty linear with the desired temperature, for example if 45% required for 150C, then for 200C you can set the dial and 60%, and be sure will be pretty close without any measurement.
So, if you preheat the machine and charge at half the maximum temperature, then the maximum power during roast could be double compared to minimum power percent. Something like 32 to 64%, just an example.
For this reason, a proportional-on-setpoint control method could be more reliable and accurate than an error based PID control.

For other machine builds, that rely more on heat accumulation, the relationship is different.

All heater percent figures relates to output power percents, not only for PAC, there are three methods to modulate power implemented in TC4.
Ironically, phase angle is the only from these three methods where heater power percent is not directly related output percent, but translated using a non-linear formula.
 
jake415
Just checked and Max_OT1 is set at 100. Still, as shown in the attached graphs the current flow maxed out starting with the artisan slider set to just 50%. That was unexpected since the fan didn't max out until 100.

When the heater burnt out, I was testing the FB machine with the fan on pretty high. It is possible that there was a short or something, but I also suspected the fact that 50% heater control here resulted in max current for the heating element. Can't be sure, but I am weary of using the heating element at max current now (it was 13.4A).

Appreciate learning that temp and power are pretty linear. Also, it's interesting to hear about how PAC related to power percent. By the way, what is an example of set point control method with artisan?
 
renatoa
Control methods are done in TC4 board firmware, via the CONFIG modes in user.h
Artisan just demands: "give me this, give me more/less", not how is this done.

That current value maxing at a certain percent is strange for me to hear... sounds like a delayed phantom zero cross signal is generated...
Please, can you switch the config mode to PAC2, i.e. heater using ICC instead phase angle, and see if it still maxing ?
 
jake415
Looks like it was already on PAC2. Should I try a different one, or change something perhaps? I'm a bit confused because in the comment after CONFIG_PAC2 on my version of the TC4 code, it says "phase angle control on OT1 (heater) and OT2 (fan)," without mention of ICC. I did see that it is mentioned though after the ifdef statement:
#define PHASE_ANGLE_CONTROL // phase angle control for OT2(fan) and ICC control for OT1(heater).

Looks to me like all of the config options beside PWM use ICC for the heater and not PAC, but I'm not sure.
Edited by jake415 on 04/27/2022 4:17 AM
 
renatoa
This is how it looks the last version, assumed official:
https://github.co...user.h#L24

Quote

#define CONFIG_PAC2 // integral cycle control on OT1 (heater) and phase angle control on OT2 (fan); ZCD required on IO2


There is CONFIG_PAC2_IO3HTR that is using fast PWM on IO3 for heater.
Such signal should go into a dedicated PWM driver, not a standard/regular SSR.

Of course, changing the code to use PAC instead ICC is not difficult.

If we clarified you are using ICC for heater, then the 50% maxing could point to the fact half of the sines are skipped, for an unknown reason... @iPa's post #64 above, about reversed polarity of the ZC signal for RobotDyn board could be a hint...
Edited by renatoa on 04/27/2022 6:50 AM
 
jake415
That was really helpful, thank you. I changed the code based on IPA's suggestion in #66 and it fixed the issue. Rising edge helped me to avoid skipping the triggers, and gave me a pretty much linear chart from 0% to 100%.

Attached are two Artisan % vs Current graphs:
One for robotdyn using FALLING zero cross trigger
And the other using RISING zero cross trigger

The falling edge trigger graph shows that current is maxed starting at 50%
The rising edge trigger graph shows a somewhat linear change in current from 0% to 100%!
jake415 attached the following images:
rising_zc.png falling_zc.png
 
progen
I can't thank all of you enough, especially renatoa, who I bugged nonstop in private messages. I finally got mine working. Now the last thing to do is to tweak the PID settings but that'll have to wait because I'm upgrading the heating element to a 6500 - 7000 watt one from 5000+. It's a one week break for some companies in my country because of the Muslim new year. I'm lucky that a bean importer had given me more than 10kg of stale beans for testing when the actual PID tweaking starts.
 
Jump to Forum:

Similar Threads

Thread Forum Replies Last Post
Hottop 2k+ Light Roast Profile HotTop Roaster 10 03/22/2022 6:06 PM
RobotDyn Transistor DC Relay Module Dataloggers/Controllers/Rate of Rise Meters 2 01/14/2021 3:41 AM
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
Roaster for light commercial work WHAT SHOULD I BUY OR MAKE 4 04/14/2019 7:45 AM
Homeroasters Association Logo, and all Content, Images, and Icons © 2005-2016 Homeroasters Association - Logos are the property of their respective owners.
Powered by PHP-Fusion Copyright © 2022 PHP-Fusion Inc
Released as free software without warranties under GNU Affero GPL v3
Designed with by NetriX