Dremel 400 failing with constant start-stop-start-stop

[Cykar]

I have a Dremel 4000 which I use when working on jewellery with a flex shaft.

Everything was fine until Monday, when after a couple of hours work it failed. I turned it on after changing the mandrel I was using and it proceeded to start-stop-start-stop as if one of the brushes in the motor was not getting any power.

I had changed the speed/motor control board last summer and everything has been fine up until now.

Taking it apart, I cannot see anything obviously wrong.

I've made short video showing the problem, and a quick look inside:
https://youtu.be/7uyt6AFn4r8

There are 3 chips on the board
The markings on the 3 pin chip are:
T4
3560
GF135 824
(a T4 3560 TRIAC?)

One of the other two has the following:
WL
2008B
830166 (could be 830156 but I think 830166, hard to read)
I have found a U2008B 8 pin chip which is a phase control circuit, applications including motor control (datasheet: https://www.mouser.com/datasheet/2/36/doc4712-29429.pdf)

The last chip has the markings:
0621
MZ915
I can't find anything on this - no idea.

Has anyone any idea what is going wrong here? Is it the TRIAC or is there something else happening?

All the diodes check out fine, but that is limit of what I know how to test...

Any help gratefully received.

[SilverSolder]

What about the brushes themselves, are they in good shape?  The commutator too? 

[Ian.M]

It could be a bad connection e.g. a broken wire.  If so, its normally making contact, but the reaction torque from the motor starting up causes it to loose contact, stopping the motor, repeat ad nauseam.

[perieanuo]

beside broken wire/bad potentiometer, check all smd's for correct pin soldering, you may have a cracked resistor or capacitor
take them out one by one (start with R and C) or just resolder all, takes 15 mins
in vibrating machines, sometimes you can't see cracked devices, it happened to me once with a resistor, with multimeter it was ok, when i put the soldering iron, he split literally in 2 pieces
but as momeone said already, start with wires/ potentiometer

[Cykar]

The brushes are fine, relatively new and the commutator seems ok, tho i would have thought that if there was a problem with it it would effect both brushes/sides of the circuit and not just one side. I have also checked the wire connections and having disassembled and reassembled several times in the last few days with the problem remaining exactly the same I am confident that those connections are also good.

I will check the other components, resoldering where i can - it's all quite fiddly tho.

[Cykar]

Well, resoldered all the components and no joy. Same problem, nothing's changed 

[Ian.M]

The next thing *I*'d do, as you've found a pinout and datasheet for the phase control chip, is extend the motor wires so I could reassemble the casing to hold all the motor parts in place and be able to check voltages, waveforms etc. while its actually running.   I'd probably strap the body down with cable ties to an improvised cradle screwed to a plank of wood, and similarly mount the control wheel so its not all flopping around when I try to test it.

Be careful, its all mains live, so is a serious shock hazard.  If you haven't got experience with non-isolated mains power control circuits (e.g, working on the primary side of SMPSUs, or on  VFDs or similar), this probably isn't going to be the easiest repair project to start with. 

Hint:  Hand-holding test probes on a live mains board is risky.  You can get away with *ONE* for brief go/no-go checks if the board layout is low density and its firmly secured so it cant move, and there's somewhere safe (and fully insulated)  to rest your hand or arm, otherwise (and always preferable if you need to study a waveform or concentrate on a reading, or if the board is difficult to secure firmly), you need to solder on wires to bring out the points you want to measure to connect to your probes.  For a fine pitch SMD board like this I'd probably run Kynar wirewrap wire through 1mm bore sleeving, to get a fine enough wire to solder to small pads without risking ripping them off the board, with enough insulation to be reasonably safe to use.  At the probe end, the female contacts from a PC IDE drive 'Molex' power connector (e.g. stripped from a dead PSU) fit most multimeter probes if you dont have an insulating cover fitted to the probe tip.  Some heatshrink sleeving over the probe connector will be required to avoid exposed bare metal.    Connecting to a scope is more problematic - you *SHOULD* be using an isolated differential probe, but if you are using two channels differentially, consider fitting BNC adapters to your probes and connecting the test wires to the center pins of a pair of BNC sockets, with their shells connected together.
In all cases the test sockets should either be securely mounted with the probe in them properly supported, or you should add insulation over all exposed metal, preferably both!

[Miti]

In your video, it seems to cycle from zero to full speed. Does the potentiometer do anything? If you turn it, does it go from zero to a lower speed?

Edit: That U2008B should have soft start, not sure if it is used in this case. It seems to control the triac, otherwise it wouldn’t turn off, but cannot adjust the triac phase. It goes full on, it then detects over current and shuts it off. Check the caps around that IC.

[perieanuo]

In your video, it seems to cycle from zero to full speed. Does the potentiometer do anything? If you turn it, does it go from zero to a lower speed?

Edit: That U2008B should have soft start, not sure if it is used in this case. It seems to control the triac, otherwise it wouldn’t turn off, but cannot adjust the triac phase. It goes full on, it then detects over current and shuts it off. Check the caps around that IC.

i suspect also the potentiometer, especially from central tap to pcb, the OP said nothing about checking this or putting some contact cleaner on the potentiometer...

[SilverSolder]

In your video, it seems to cycle from zero to full speed. Does the potentiometer do anything? If you turn it, does it go from zero to a lower speed?

Edit: That U2008B should have soft start, not sure if it is used in this case. It seems to control the triac, otherwise it wouldn’t turn off, but cannot adjust the triac phase. It goes full on, it then detects over current and shuts it off. Check the caps around that IC.

i suspect also the potentiometer, especially from central tap to pcb, the OP said nothing about checking this or putting some contact cleaner on the potentiometer...

I was just going to say the same thing.   Also, check the off/on switch itself.   -  The symptoms still "feel" more like some kind of mechanical problem, more so than something electronic, just yet.

[Cykar]

Don't think it's the pot, as it seems to be doing what i'd expect it to do - the pot seems to me to be adjusting speed through speeding up the phase change if that makes sense?

The switch also is fine and doing what it should.

It's funny to me (not the right word but you know what i mean), that all was completely fine then, without any change leading up to it, failed. So, changing speed was doing exactly what it should, no change at all, and the on/off switch, similarly, was working perfectly, and still does in that it starts up, or tries to.

I'm wondering if there's something mechanical I've missed somewhere.

I can remove the board and test the leads while plugged in but not connecting to the motor itself, that might tell me something?

Wondering if there is a broken connection somewhere on the motor itself, maybe under the plastic housing where the leads plug into the coil? But that housing is attached pretty rigidly so might be difficult to get it off or get under under it.

Not sure I'm up to attaching extension wires to a surface mount chip and checking what's coming out while strapping down the motor and keeping my tongue at the right angle - I can just see disaster waiting round that corner


I'll try again next week when I have some time again.

[SilverSolder]

Don't think it's the pot, as it seems to be doing what i'd expect it to do - the pot seems to me to be adjusting speed through speeding up the phase change if that makes sense?

The switch also is fine and doing what it should.

It's funny to me (not the right word but you know what i mean), that all was completely fine then, without any change leading up to it, failed. So, changing speed was doing exactly what it should, no change at all, and the on/off switch, similarly, was working perfectly, and still does in that it starts up, or tries to.

I'm wondering if there's something mechanical I've missed somewhere.

I can remove the board and test the leads while plugged in but not connecting to the motor itself, that might tell me something?

Wondering if there is a broken connection somewhere on the motor itself, maybe under the plastic housing where the leads plug into the coil? But that housing is attached pretty rigidly so might be difficult to get it off or get under under it.

Not sure I'm up to attaching extension wires to a surface mount chip and checking what's coming out while strapping down the motor and keeping my tongue at the right angle - I can just see disaster waiting round that corner


I'll try again next week when I have some time again.

An open circuit or intermittent connect to a winding is a possibility.  You could try spinning the motor manually and see if it "catches" at certain points?

[perieanuo]

Don't think it's the pot, as it seems to be doing what i'd expect it to do - the pot seems to me to be adjusting speed through speeding up the phase change if that makes sense?

The switch also is fine and doing what it should.

It's funny to me (not the right word but you know what i mean), that all was completely fine then, without any change leading up to it, failed. So, changing speed was doing exactly what it should, no change at all, and the on/off switch, similarly, was working perfectly, and still does in that it starts up, or tries to.

I'm wondering if there's something mechanical I've missed somewhere.

I can remove the board and test the leads while plugged in but not connecting to the motor itself, that might tell me something?

Wondering if there is a broken connection somewhere on the motor itself, maybe under the plastic housing where the leads plug into the coil? But that housing is attached pretty rigidly so might be difficult to get it off or get under under it.

Not sure I'm up to attaching extension wires to a surface mount chip and checking what's coming out while strapping down the motor and keeping my tongue at the right angle - I can just see disaster waiting round that corner


I'll try again next week when I have some time again.

think again, if pot central tap makes intermitent connection, this is exactly the result
contact ok>> set speed, contact interrupted>>min or max speed, depending on schematic design

[joeqsmith]

Did you get your hand with an aggressive cutter bit installed?   

[tanveerriaz]

can you try  bypassed the speed control.

[Cykar]

The pot you can feel as you turn it, meaning i think it might be a rotary encoder rather than potentiometer.

I have established as far as I can that the coils are good, and that the wires connecting them to the board are also good.

I'm out of ideas and have reached the limit of my knowledge and ability to test.

Thanks for all the suggestions, but I think I'm going to have to admit defeat on this one :-(

[rh100605]

The devices are the  U2008B speed controller from Atmel  and the other 8 pin ic is an ST 062 dual op amp (marked 062 I not 1)

The pot value should be  10.0k also check the electrolytic capacitors..........

More pictures on /www.elektroda.pl/rtvforum/topic2925200.html#19599893

[rh100605]

Hi,
What did you do with this Dremel 4000 ?

[Cykar]

@rh100605 thanks for your reply!

I've still got it. when i've got some time i'll give it another go.

[mcinque]

Is this dirt/flux/coating or it's blown?

[rh100605]

The picture of the 062 I shows a blown device . No idea why it went but it is a fragile dual jfet opamp....

[rh100605]

Hi,
I have a 220volt Dremel 4000 here that lacks torque and stalls easily. I have checked out the motor and it is fine.
To examine the electronics I have attempted to draw the circuit diagram in the attached pdf . Maybe it will help you  as well. If you spot any mistakes please let me know.
Some components such as capacitors and diodes I have been unable to identify. Component numbers are those on the board.
My first change will be the  ST T435-600 Triac  as there is no sign of burning/damage anywhere and I have measured all the resistors.

UPDATE: Well despite the armature reading 6.9 ohm between all adjacent segments + >10M to metalwork and having cleaned out the segment slots; when I short out the Triac the motor runs very hot and at medium speed, no wonder the thermal fuse had gone before! Not the electronics then...
Replacing armature fixed it.... I now have spares for the circuit board in my useful bits box.

[Cykar]

Looking at this again, and having changed the dual op amp with not change, I am now thinking it's the triac that's at fault.
Don't quite understand enough about how they, or this particular circuit works, but from what I understand it switches which way the current flows, and, thanks to @rh100605 schematic, I reckon the triac is responsible for powering the motor in 2 alternating phases, but that it is only working in one direction - that is my latest working hypothesis anyway.

All the diodes and caps measure fine as far as I can tell, and it's not as if it is not working at all, just that it keeps pulsing rather than running continuously.

Will get another triac, and if I can get the original off, which might be hard as it's back plate is soldered to the board and I don't have a hot air station, then I'll replace it and see if that works. Can't think what else could be going on. I bet bigclive could tell me 

[rh100605]

Before replacing the triac, try shorting the input and output (A1/A2 )of the triac and see if the motor runs at high speed continuously
This is easy to do as the triac pin to the outer of the pcb should be shorted to the middle cut off pin (solder bridge?). This leaves the gate open...
Then reassemble the tool and switch on/off at the mains with the tool switch on. This experiment will eliminate the motor wiring + thermal fuse except for the control/feedback wire that connects to one of the brushes with a phosphor bronze tab.

If the pulsing goes away then my guess is not the triac but the speed control circuit starting up, not being able to see a feedback signal and then shutting down. It then repeats. Your video shows even that the speed control works to control the speed for about 1 sec before shutting down and trying again. The first target for me would be the electrolytic capacitor that smooths the power for both chips. They have a finite life.
Also check the connection from brush casing to board for the feedback wire, if open the motor would power up and power down...

[Jeff eelcr]

Several years ago I had 2 that did the samething problem was the armature, at speed the winding would seperate at a connection (epoxyed) microscopic crack.
Jeff