Reverse engineering of a mains power controller.

Test and investigation of a phase angle control motor speed controller from China.
There are two components on this PCB that cost me MORE to buy individually than the entire cost of this assembled unit including shipping!
It's a very standard phase angle control module for varying the duty of mains powered devices like lamps, transformers or some motors (mainly universal motors).
The circuit does have basic suppression, but lacks the important choke that prevents strong radio interference, especially on older AM radios.
Other than that it's a nicely built module using a textbook design and a beefy triac on a chunky heatsink. They claim it's rated for up to 2kW, but I'd be wary of using it on anything near 1kW continuously.
Note that the triac used has an isolated tab but that's not always guaranteed, so the heatsink should be regarded as being at mains voltage, as should the whole PCB. If the module is mounted on a metal panel (using the potentiometer nut) then the panel should be grounded/earthed for safety.

[gskempe]

What a nice presentation and reverse engineering job. I just purchased one of these and was about to reverse engineer the circuit when I came across your presentation. You did an excellent job, saved me some time, and gave me a better understanding how the circuit works. Thank you very much. Respectfully, Steve

[Harkendown]

Hi. I'm no electronics engineer but I do have a need for these regulators to control motors of various woodworking tools I am building. I have bought 10 of these to date, from various suppliers and prices from 1.50 to 3.5 gbp. Only two of the ten do what they say on the tin. The other eight have output voltages from 200 to 240 (not 50-240 as promised). I found your video very edifying and as a result of what I learned went back to re-test the faulty ones to make sure I was testing under load and with minimised potentiometer setting.
All I can say is you were lucky to find one that worked! Thanks for the video.

[KarldorisLambley]

me again. i just bought some diacs to go with the triacs i bought ages ago. thanks mate, as i keep saying, your work has transformed my life.

[johnjurotte]

DERP??
That's a word that we don't use in Australia. I will add it to my vocabulary.
Thanks again for another great video

[cornelisspronk]

I have used this to control the heating element in a homebuilt coffee roaster from a 230 V 1500 W paint stripper. To roast the coffee it runs for about 12 minutes at 650 Watts. The BTA-600 has failed after several months. I have replaced it with a BTA-800 a couple of years ago, and it is still going strong after using it at least every other day. The line voltage here ranges from 230 to 240 V. The ever changing line voltage is enough to mess up the roast, so I needed to tweak it. I removed the 1 Meg Potentiometer and replaced it with a 50K potentiometer in series with a 320K fixed resistance. This allows me to keep the temperature in a narrow range for a satisfactory roast.

[Peter-pubo]

Hey big Clive, I can't thank you enough (well I will see if I have you in my patreon list) there is so much rubbish out there and your DETAILED explanation helped me so much to solve my problem.
Cheers and all the best to you stay healthy.

[faxepl]

Went to Ebay, bought 2 of those while listening to you talking :D

[GKVVx]

Your analysis is very good and helpful for needy repairs if any. I purchased recently one from China. It is working good for my mini drill (220v, 350w 2600 rpm) which I use for day to day minor works and also for my Bosh angle grinder 600w 11000 rpm (to the reduction of speed upto 50% ). Any way my other Drill Bosh 13 mm has got builtin speed control. Thank you. With regards - VG Krishna, from India.

[carmelpule]

For resistors in parallel
1/R= 1/R1+1/R2
Hence 1/R-1/R1= 1/R2
(R1-R)/R.R1 = 1/R2

Hence R2= (R.R1)/ ( R1-R)
R2 = ( 406, 000* 2, 000, 000) / ( 2, 000, 000 - 406, 000)
R2= (406*2000)*1000/(1594)
R2 = (812, 000) / (1594)
R2= 509, 000
So the control resistor is a potentiometer of 509, 000 ohms 470, 000 was close enough. A good logical calculation by the narrator. Well done.

[KarldorisLambley]

i found triacs confusing. mainly because of the all the names i used to mix up - triac, variac, varistor, thermistor, thyristors, thyratrons, and klystrons. i know they are all very different things. now. this vid was fantastic, during which, i made copious notes. i have finally understood triacs. the circuit made it all crystal clear. i want to buy a box of them now to play with.

[alastairbarkley]

It's a single phase shift triac dimmer which will give erratic operation at low triac conduction angles - a nuisance for lamp dimming, more serious with motor control. The so called 'hysteresis' effect. The 100n capacitor suffers very heavy loading as the voltage across it collapses when the triac fires. Solution - a double phase shift circuit, which involves only two additional components, a 1/2 watt resistor (typically 15k for both 120v and 240v supplies) which connects between the junction of the existing 4.7k resistor and 100nF cap and the left hand diac terminal. The additional 100n cap goes between the left hand diac terminal and Neutral. This greatly reduces loading on the original cap and improved operation at 'dim' or slow speed triac settings. The second 100n cap needs to be prooerly voltage rated - same as the first one. This is a bona fide mod found in all the thyristor books.

[Firecul]

From what I remember (I haven't done this for 9 years now) the knob pot looks like 510kΩ.

[minkorrh]

Excellent info! I have a few of these kicking around and was just going to stuff them into a plastic box to use as a hacked up soldering station and hot air gun. I was wondering about the potential for electrocution when i noticed that 2 pins of the pot are directly attached to one side of the outgoing AC. Thanks again for another entertaining/informative video. Esecially enjoyed the 'Hitachi' teardown. Had to have a drink to Lily the Pink after that.

[alastairbarkley]

Thank you Clive. Pedant point but the snubber network you mention is not 'a filter'. It is indeed to prevent false triggering of the triac with inductive loads. Basically, the triac itself has an internal capacitance (gate to MT). In an inductive load, the voltage across the 'inductor' is not in phase with the current flowing through it. If the current flowing is zero but the voltage is high, the gate 'capacitor' can acquire sufficient charge to trigger the gate (and hence falsely trigger the triac). The faster the change in voltage (dv/dt), the greater the risk. The snubber network is simply a series RC which reduces dv/dt.

[williamsquires]

No, it's closer to 500k: (2M x 500k) / (2M + 500k) = 400k. Which is closer to the 406k, than if you use 470k, which gives: (2M x 470k) / (2M + 470k) = 380.56k.

[abousamah]

Thank u very very mutch
Your explnation is enjoying and very simple
And your attachment schemes make the episode very nice
Thanks

[Tocsin-Bang]

I built a speed controller using this same circuit a few years back Sadly its died for some reason (probably not a big enough heatsink), but I think I'll buy one of these and just add either a small fan or a bigger heatsink.

[harviecz]

Replace pot with photoresistor, heatshrink LED to it, connect led to arduino. BAM you have PWM control for universal motors. Add another LED and phototransistor to read RPMs from motor, write some PID code, BAM you have closed loop feedback motor control!

[geofflotton]

The description says it all. " This controller uses novel bi directional high power SCR. It can be very convenient adjust the current up to 25A and solve the over current problem of the heating wire resistance is too small in the case of cooling causing good.

[anthonyshiels]

The formula for calculating a number of resistors in parallel is given by
1/R = 1/R1 + 1/R2 + 1/R3 + ...
Series capacitance has the same format.
With a University Degree in Physics and Chemistry I never used these formulae since I was in secondary school which I graduated from in June 1974!
I had to check them on Google.