Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Handdrill trigger switch control

Status
Not open for further replies.

Blueteeth

Well-Known Member
Hey there,

Well I have a cheapy black & decker KR504 500W hammer drill. Its virtually new and working fine but.. I wish to use it in a drill stand for half decent holes.
Its runout is shocking and its over powered, but right now its the only option I having for drilling aluminium for a project (cast aluminium, M4 bolt holes, 9mm depth).

Because like many hand drill the speed control is via the trigger switch, and the only way to keep it on hands free is via the hold switch (which holds the trigger fully on) it can only be used at full speed, which is LOUD and over kill - especially when countersinking aliminium.

So, I decided to see if I could modify the trigger circuit to reduce the speed when the trigger is fully depressed (and held by the button). Saw a basic SCR circuit there, but I was surprised just how simple it is... SCR (600V 12A), a 150nF.. NO diode on the gate , a fix thick film printed resistor, and a variable resistor track. Along with a suppression cap across the circuit that is all.

Do you think if I replaced R1 (sets the minimum value of the trigger) with a 160k in series with a 1Meg pot, I could 'set' the maximum power? As this will be connected to live, despite the low current, I'll have it as a plastic housed preset pot inside deep inside the drill, with a hole to adjust it with a plastic screw driver. With the added pot at 0, it should behave the same (minimum 160k resistance), but when adjusted it should add more resistance (increasing phase shift of the power waveform) and in effect simulate releasing the trigger a bit slowing it down.

Is it this simple? or am I missing something... I attached a schematic, values are measured - except the SCR which is MCR12DSM. The resistors are all printed thick film, but the resistor in parallel with the carbon track... is an odd shape (makes a difference?).

If anyone can chime in, I'd be grateful :) I'm afraid I'm not overly well versed in SCR circuits (aside form basic triggers).
 

Attachments

  • B&D_drill.png
    B&D_drill.png
    16.9 KB · Views: 165
Not an electronic solution, but couldn't you use a wire tie to set the speed you want. Wrap the wire tie around the trigger and handle then tighten to a set speed. A cord with a toggle switch and plug for the drill would give a fast way to turn it on and off, to keep the speed set.
 
Think you're right Ron, although its definitely a SCR. Its not easy to trace it all out because the connections to the motor, AC line, and snubber, are inside the trigger (molded plastic), so I just did a quick continuity job :/ But the schematic I posted is of the main board, a hybrid PCB with a heatsink on the back, and copper fingers for connections to the trigger.

Shortbus, you're right, I could just use a cable tie, but I'm unsure of how stable that would be. In fact its a damn good idea and I may try it (after I've sorted out the runout a bit... they have the main bearing mounted on plastic studs...ugh). Good call!
 
I it was me I would just wire it full on and use a common heavy duty light dimmer for the on/off and speed control.

Most light dimmers have no problems with working as speed controllers on the brush type motors.
 
I've seen this done where the owner replaced the variable resistor trigger pot with a photocell and lit that with an led driven by pwm.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top