ink bird temperature controller mystery - part 2

 

When I decided to write about the controller and title it a mystery because I had reservation of what was represented on the product page. It is one of those thing that taxes your ability to process ambiguous and conflicting information.

I live my life to be generally cautious, and guard myself from fools. When I received the controller I checked the connection descriptions on the unit, and read the manual carefully. Everywhere suggest this is a 12V version of the ITC-2000x family of products. I checked Amazon's product page not once, but at least three times and it sure describe this is a 110V version.

it clearly states this is 110V

Since the manual and the unit says 12V I did all my testing with 12Vdc. As I was left with the possibility that this thing does not care whether you feed it anything from 12Vdc, 12Vac, and up to 110Vac, I was planning to test out on 110Vac last. In case I am wrong the will destroy the unit after all the tests I planned to conduct. I was very apprehensive that is can be powered from 12Vdc to 110Vac. My skepticism turned out to be justified.

First to the good. Using 12Vdc supply. The unit is everything I expect. The temperature trigger spread of 3 degree F is adequate for my application as long as I choose the sensor location wisely. I like the compressor lockout delay of 0 - 10 minutes. I wish it give fractional minute increment but it is only settable for one minute increment from 0 to 10 minutes. It's industrial design is also tasteful and fits nicely with the Westfalia decor and aesthetics. 

It will also remember the last power on/off stage if I shut remove and restore power. That means I don't have to switch on the air conditioner with two on/off switches, as long as I leave this controller On when I switch off the AC's master power.

I conducted good amount of tests with the Coleman AC

With this controller I have great latitudes in controlling the cycling characteristics of the AC, as well as any temperature set point I wish, within reason.

After I was totally satisfied with its operation, I started planning on where and how to mount it. But since I want to use the AC's master power switch to turn on this controller also I want to power it with 120Vac (same as 110V). I was not dumb enough to try it with it hooked up to the air conditioner circuit I built in case something goes boom!

I unhooked it from the air conditioner and bring it back to the bench to perform this test. I wired it up and plugged it onto a power strip with the switch set to off. I stood back and switched on the power. Immediately something pop inside and a billow of smoke. The mystery solved, but unfortunately I had to destroy it to find out.

I am disappointed with this to say the least. I went and order the ITC-2000 which clearly states it is the 110V version. I actually order another brand that is cheaper and I know there are identical to the correct Inkbird ITC-2000. I am keeping the fingers crossed that the right one will retain the previous On/Off state when power is cycled.

this other brand call the model STC-1000 and there are three versions; I know this time it should be unambiguously correct

Onto the postmortem. I want to see the design and construction of this little gem. It is well designed to be cost efficient to manufacture, especially the housing. Everything snaps together with no need of a screw.

everything snaps together without the need for one single screw on the assembly

one of the power supply wire got blown off the PCB; the capacitor also popped; note that the main PCB has the build option for a step down transformer (the square symbol where the broken wire) making it the 110Vac model

Being such high EOS of 100V there are little left of the semiconductor devices not damaged. There ain't no repair for this thing.

even the bottom side of the PCB was exposed to high energy of EOS

I am pleasantly surprised it has real button switches rather than printed conductive ink contacts

a single A/D flash MCU forms the heart of the controller

There is very little on the board as a single Holtek HT66F40 A/D flash MCU forms the heart of it. The remaining parts are the segmented LED display, beeper, relays, one linear Vreg, transistors and other discretes. Holtek is a Taiwanese company.

Reviewing the Holtek MCU it is no surprise like any other contemporary counterparts there are huge number of unused resources and capabilities that this temperature controller does not utilize. It can be many fun project only if I can avail time to hack it, by preprogramming it. By reprogramming I mean gaining access to the MCU programming and rewriting the codes while using the existing hardware design. This is unlikely possible without de-soldering the MCU and replacing it with another one as the existing one is likely locked or with one time programmable ROM.

This is the same thing with the Microchip MCU in the Dometic Heki panoramic skylight. Only if I can avail the time to tackle that project.

So after inspecting the design and construction I decided to buy a 110Vac and a 12Vdc versions each. I can always use a DC powered version for my projects.