Triac speed controller for handheld drill

[gaminn]

Hi,
I would like to control speed of my AC 230V handheld drill. I already purchased cheap chinese triac controller https://www.tipa.eu/en/dimmer-and-speed-controller-for-commutator-motors-up-to-4000w/d-245065/ . It changes speed of the drill but for low speeds the drill has almost no torque, therefore it is useless.

I discovered an old thyristor controller made by my father 30 years ago which perfectly controls speed of the drill in low range as it has EMF feedback from the motor, unfortunately it is thyristor based so it is only half wave.

Do you know of any ready to use triac controller which can be purchased on eBay, Amazon, AliExpress, etc which can control speed of my drill in low range (i.e. it has EMF feedback)? Thanks

[DavidAlfa]

Are you sure it's only half wave?
Thyristors can be used with rectified (Unfiltered) ac.
If not, you could add a full bridge rectifier before the scr driver.

Also you could modify the triac circuit:


Source:
https://www.homemade-circuits.com/how-to-make-versatile-closed-loop/

[gaminn]

It is this circuit:

(I'm sorry, I can't add inline image to my message, it doesn't work, please open the thumbnail below)

It features back-EMF feedback, therefore it maintains (almost) constant speed under load. I think rectifying input power supply would break the circuit.

[DavidAlfa]

Yeah, adding a full bridge rectifier should work
The + should go to L1 side and - to L2.

[gaminn]

Really? I'm not expert on these analog power circuits, but would't rectifying break the back EMF feedback, because for the first half wave the feeback will be as expected but for the other half (rectified) wave the feeback will be 180° shifted (as back-EMF is still alternating)?

[Benta]

You must be talking about a different circuit, because the one you show has no BEMF sensing.

[gaminn]

Not back EMF sensing directly but it compensates for loss of speed under torque. It is widely used circuit in our country and it was first published in 1989 in local electronics magazine. I tried it, it really has no problem to run my drill at very low speed and still has enough torque to drill.

[floobydust]

For constant torque I see current-sense used on the motor speed controllers, for power tools and sewing machines even.
This thread has the Atmel U2010B https://www.eevblog.com/forum/reviews/2021-craftsman-bench-jointer-cmew020-help-me-reverse-engineerhack-this-please/

Silicon Chip Magazine has done a few universal motor speed controllers over the years, you can dig for the full circuit.
Full Wave, 230V Universal Motor Speed Controller - March 2018
10A Universal Motor Speed Controller, Mk.2 - February 2009
Speed Controller For Universal Motors - October 2002

Not a lot of room inside a hand drill for electronics.

[gaminn]

Thank you very much. I already noticed U2008B. However, it is obsolete chip. Only Chinese sellers have it and it is probably not original. I guess U2010B is the same story.

I have plenty of room inside my drill. Actually it is not handheld drill, it is cheap magnetic drill from AliExpress, but a regulator for handheld drill works for it very well.

[IanB]

I know that this exists: https://www.harborfreight.com/router-variable-speed-control-dial-59386.html

Since it is probably not exclusive to Harbor Freight, you might find a similar version that works on 230 V by searching the usual places like AliExpress.

I cannot tell you if it has any kind of feedback, but it might do since it is designed for motor applications.

[strawberry]

https://www.tindie.com/products/sakurawizard/triac-power-regulatordimmer-25a-jr20/
simple triac phase angle regulators can run ~50...100% and dont accept loads below ~50W . need modified circuit or dedicated IC

[gaminn]

I know that this exists: https://www.harborfreight.com/router-variable-speed-control-dial-59386.html

Yes, I found it on AliExpress but also found there a review saying that it didn't work for (probably some particular) motor and it is mainly suitable as a dimmer.


I got full article and schematic for https://www.siliconchip.com.au/Issue/2018/March/Full+Wave%2C+230V+Universal+Motor+Speed+Controller and I like it There is ASM code available for the PIC. If there are no known drawbacks of this circuit I'll give it a try.

[gaminn]

I made a prototype of https://www.siliconchip.com.au/Issue/2018/March/Full+Wave%2C+230V+Universal+Motor+Speed+Controller and it seems the controller works as expected. Not enough time to test it however, I'll see if something burns later - some drilling planned on weekend.

[gaminn]

I noticed that when the controller is powered on the drill connected to it sometimes starts rotating for fraction of a second.

This is the schematic: https://archive.org/details/silicon_chip-2018_03/page/36/mode/2up

The problem is not caused by the microcontroller because it happens also when Q1 (BC337) is disconnected from G of TRIAC1. I tried to increase value of the capacitor and decrease value of resistors in the snubber circuit accross A1 and A2 pins of TRIAC1 to decrease dV/dt. It didn't help.

Maybe bad triac? Because it is made by an unknown factory, at least I don't know it.

Or isn't here any flaw in triggering the triac in the circuit that can cause this issue?

Thanks for any help.

[ozkarah]

Do you have a scope? You can try monitoring the output pin of the microcontroller and the gate pin of the triac. 

[gaminn]

Yes, I have a scope but as I already wrote, the MCU is disconnected from the triac, or in other words I left gate of the triac floating or I also tried to ground base of Q1 so there is no way it can turn on. Nothing changed the situation, even with the gate of the triac floating, it sometimes happens.

[ozkarah]

Would you mind pulling the triac gate to the ground and trying again?

[gaminn]

If I pull triac gate to ground (which is e.g. pin 8 of IC1 or emitter of Q1) then I think the triac will always be on, as there is current path formed from +5.1V to ground via A1 and G terminals of the triac.

[ozkarah]

Sorry, you are right. I hadn't checked the schematic.
Since you tried pulling down the base of Q1 triac shouldn't switch on in that case.

[gaminn]

I think I have bad triac with low dV/dT rating. I will replace it with T4050-6PF snubberless triac, min 1000 V/us. When zero-cross SSR relay is used to power on the controller then I can switch it on and off as how many times I want and the drill never rotates momentarily at power on event.

[gaminn]

I found that I don't understand how +5.1V is created in this circuit: https://archive.org/details/silicon_chip-2018_03/page/36/mode/2up   They write there: "When the Neutral line is positive with respect to the Active line, current flows via the 470nF capacitor and diode D1 to the 470uF capacitor" My question is: Why would current flow via D1 to the 470uF capacitor next to ZD1, if there is no path the current can return to Active line?

Also, if I measure voltage accross ZD1 on my board there is 5.6 V, and accross ZD2 there is 4,9 V. Both zener diodes are not saturated which I think is not very good. I think the only reason there is some voltage on +5.1V rail is due to 47 ohm resistor connected between +5.1V rail and A1 of TRIAC1 which is effectively Active in line. Then there is at least some path (via 47 ohm, ZD2 or 470uF, D2, 1kohm and finally 470nF) the current can flow between active and neutral line.

Any explanation on how the circuit works welcome, thank you.

[IanB]

This is interesting. I am far from sure of the answer to your question, but it almost seems as if there if the active line should be connected to circuit ground, and that there is a missing symbol on the schematic. It almost seems like the whole circuit should be referenced to the active line potential, and that the microcontroller is measuring the voltage difference on the neutral line relative to this (I have not had time to read the text to verify this). But the annotation of "zero volts detector" on GP2 seems to imply this.

If the active line were connected to circuit ground it would provide the current return you are looking for. If this were the case, the text would be saying that the whole circuit needs to be insulated and isolated from the mains protective earth ground. It is not unknown for circuits to have a live chassis if they are properly enclosed and insulated.

[floobydust]

Always check the article errata.

"Full Wave 10A Motor Speed Controller, March 2018:
(1) The mains Active wiring to the fuse holder shown in Fig.2 should show the incoming mains wire (brown) connecting to the tip of the fuse holder rather than the side ring terminal. The wire through the current transformer should then connect to the side ring terminal of the fuse holder. Also, the paragraph before the “Current Feedback” cross-heading on page 39 refers to pin 2 connecting to the 4.7nF capacitor. It should say pin 5. These errors have been fixed in the online edition. (04/18)
(2) In the circuit diagram (Fig.1) on page 36, the “Active In” wire from FUSE1 is shown connecting to the top-most terminal of CON1 and it then goes through the core of transformer T1. This is incorrect; the wire from FUSE1 goes directly to T1, then to CON1 and on to the A1 terminal of TRIAC1. The overlay and wiring diagram (Fig.2) on page 40 is correct. (02/19)
(3) A bug in the original software (1010218A) prevents the feedback speed control from working after the soft-start period. Revised software (1010218B) is available online. In the circuit diagram (Fig.1), the electrolytic capacitor connected to the junction of the 10kW and 1kW resistors to the left of diode D5 should be 10μF, not 100μF. Some PCBs sold may also show this capacitor as 100μF but the overlay diagram, Fig.2, shows the correct value of 10μF. VR2 is also shown as 1kW in the circuit, but should be 10kW as in the parts list and overlay. (12/19)"

[gaminn]

Always check the article errata.

Thanks, I didn't know about their errata. However, there is nothing related to +5.1V rail in the errata.

it almost seems as if there if the active line should be connected to circuit ground

If the active line is connected to circuit ground, then +5.1V line is almost short-circuited via 47ohm resistor (between +5.1V and A1 of TRIAC1). Not sure if that is acceptable.

If the active line were connected to circuit ground it would provide the current return you are looking for. If this were the case, the text would be saying that the whole circuit needs to be insulated and isolated from the mains protective earth ground. It is not unknown for circuits to have a live chassis if they are properly enclosed and insulated.

I don't understand why there should be no protective earth ground once the digital circuitry's reference potential becomes Active in. Can you please elaborate more?

[gaminn]

I noticed there are actually 2 paths current of +5.1V circuitry can return to Active line.

If 47ohm resistor (between A1 and +5.1V rail) is removed then still the +5.1V rail is active. I noticed that the connection between C of Q1 and G of TRIAC1 has to be interrupted, then +5.1V rail have zero volts.

Either I miss something or the circuit is not properly designed as I think it depends on some weird return path for +5.1V rail's current.