Thread: Mars Lathe DC motor conversion.

Hi All,
Rather than this getting lost in the Wow huge price difference thread i am starting a new thread.

I have a 1.5 hp, 180v DC treadmill motor that i will be using for the conversion. The controller will be a KBIC240 DC (thanks Rob"nearnexus"), with a simple tacho running off a hall effect sensor.

The one undecided fact is still switching. To reverse the motor a switch needs to be put between the controller and motor, but the AC supply to the controller must be cut first. Currently i have a rotary on off switch and a toggle for forward and reverse. Basically, i don't like it! At the moment i have 2 ideas in the pipeline (i may use both yet) I am basing my thoughts around threading as this is really the only time i have ever needed to reverse the lathe.

Thought 1, use a 2 position 3 phase switch(which is basically 3 separate switches) I can wire the AC to one switch, then the DC to the other two switches to give me forwards and reverse. I am very slightly concerned about switching both the AC and DC simultaneously, but it stands to reason there will be no DC load until the AC contacts have touched, in which case arcing of the DC contacts should not be a problem. I like this idea BUT it means when threading, you can come up to a shoulder fast, turn the speed down and then you have to MOVE YOUR HAND TO THE ON OFF SWITCH.........

Thought 2 is to replace the 5k speed pot with a twin 5k pot. Then a can use the second pot as a voltage divider and use it to switch a relay on the AC line (after a master on off switch) This way i should be able to cut threads fast, slow the machine down and then stop it without taking my hand of the speed pot. The lathe can them be reversed, turned back on with the pot, repeat.

Sorry if that got a little long winded.....anyone got any thoughts as to which sounds better?

Now for some pics of the motor and fan. I am going to trial with only one. I tried to but found very little difference in a smoke test (not being a smoker i finally found a good use for incense......) A small ally frame to hold and seal the fan to the motor and it all went together a little too easily (gulp...)

(Should we tell him the thing he is trying to fit the motor to is actually called a mill?)
Can you put in say a foot switch with a relay that will cut the power when pushed, perhaps even with some sort of load so you get dynamic braking?

Michael

G'Day,
When I fitted the D.C. motor to my lathe I had 2 priorities, 1 was to have no unnecessary high D.C. voltage cabling running around the machine and 2 to have a no volt dropout as my area has fairly regular power outages and the thought of a machine restating when power is restored quite frankly soils my undies, I use separate circuits for forward and reverse selected with the original forward/off/reverse switch, a 24 volt power supply operates self latching control relays via momentary push buttons, the control relays in turn switch 60 amp contactors, the bonus is that when the off position is selected the controls are dead and there is no chance of accidental starting.
Regards,
Martin

Using the older K&B thyristor switched speed controllers of yore (~1990) supplied by Baldor just involved putting a reversing switch on the DC 180V output and then a 3A 1000V back EMF diode across the motor terminals at the switch. The diode had to be sized with a suitable current rating, 3A was fine for a 1/15th HP motor and the diode also gave active braking to motor with it's inertia from the belt driven flywheel when the power was cut. I have no idea how the current model KBI controllers are made these days or even if they bear any circuit similarities to the older controllers which rectified AC mains and then fed it into a SCR based circuit. You would need a beefy diode to brake the inertia of a lathe gear train and chuck I would expect, maybe 100 Amps. A 180V motor will readily blow a 400V back EMF diode so 1000V is a minimum voltage rating.

Edit: the diode is not across the motor terminals but on the speed controller side of the reversing switch, so it's always the correct polarity WRT to the speed controller, so braking only occurs when power is cut to the controller and the motor winds down.

Originally Posted by Graziano

Using the older K&B thyristor switched speed controllers of yore (~1990) supplied by Baldor just involved putting a reversing switch on the DC 180V output and then a 3A 1000V back EMF diode across the motor terminals at the switch. The diode had to be sized with a suitable current rating, 3A was fine for a 1/15th HP motor and the diode also gave active braking to motor with it's inertia from the belt driven flywheel when the power was cut. I have no idea how the current model KBI controllers are made these days or even if they bear any circuit similarities to the older controllers which rectified AC mains and then fed it into a SCR based circuit. You would need a beefy diode to brake the inertia of a lathe gear train and chuck I would expect, maybe 100 Amps. A 180V motor will readily blow a 400V back EMF diode so 1000V is a minimum voltage rating.

Edit: the diode is not across the motor terminals but on the speed controller side of the reversing switch, so it's always the correct polarity WRT to the speed controller, so braking only occurs when power is cut to the controller and the motor winds down.

Hi Graziano,

So the protection diode will always act as a short for when the motor generates the current on run down but will never be seen as a short to the output of the controller? So, regardless of which way you are running the motor, the diode will have correct ploarity to do this?

If so, then I think this may be a solution to an issue raised in my thread on VFD's (I digress and talk about DC motors too!)

Cheers,

Simon

Originally Posted by simonl

Hi Graziano,

So the protection diode will always act as a short for when the motor generates the current on run down but will never be seen as a short to the output of the controller? So, regardless of which way you are running the motor, the diode will have correct ploarity to do this?

If so, then I think this may be a solution to an issue raised in my thread on VFD's (I digress and talk about DC motors too!)

Cheers,

Simon

Hmm, I've been thinking about the original setup which was made by someone else for about 400 or so machines: it was a K&B speed controller which rectified the input mains 240V and then fed via an SCR/DIAC trigger circuit to the DC motor. The original controller was not designed to be reversible and would typically get fried if you just fed the controller output into a reversing switch and then to the motor as the end users would hit the reverse switch at full speed instead of setting it to zero and then reversing. The motors were belt driving a cast iron flywheel and the motors did go overspeed to 3500 RPM, so there was a lot of stored energy in the flywheel to bring to a stop.

The solution was to fit a 1000V rectifier diode before the reversing switch on the controller side (always the correct polarity), after which the motor could be reversed at full speed. 400V 1A (1N4004) diodes died after a few reverse to firsties, 400V 3A (1N5404) diodes lasted a few days of use and 1000V 1A (1N4007) lasted a few years and 1000V 3A (1N5408) have yet to fail with some units in the field lasting 25 years. The motors were 180V 1/15th Hp and used to generate a fair bit of spikes on the CRO around the 400V mark, I assume from the brushes sparking.

I have an old controller unit kicking around somewhere and it was a fairly basic SCR unit with a second bridge rectifier for use with field winding DC motors, they were a fairly good unit but are now spontaneously blowing up as they get to the 20+ year mark, they do use a lot of non 240V rated components like 1/4W resistors and 100V capacitors.

Edit: I'm not sure a diode would be effective on those modern KBIC controllers judging by the photos of the unit they have a lot going on, a diode could interfere with their function.

Hi Graz,

After our conversation RE: EMF diode I have been looking at what is available. Not much in the way of 6 - 10A 1000V diodes! I guess the voltage rating is the most important as you can always double the current rating by putting 2 diodes in parallel. Do they need to be very fast acting or would any rectifying diode do the trick?

So, in theory if you use a big say 15A 1000V diode (if you can get such a beast. Happy to solder 5 x 3A in parallel) you should be able to bang the control lever from Forward straight to Reverse and not hurt anything? Surely there has to be a catch!? Would the current generated from the motor somehow interact with the current from the controller?

I'm thinking that when you drive a motor and then cut the power and let it coast down, the current the motor produces when coasting will flow in the same direction as the current that drove it? If so, then this current will want to flow against the current path from the controller, once you switch from forward to reverse. If the motor produces an EMF of less than the voltage produced from the controller then I assume the controller will overcome it? If the EMF is greater, would that mean a second diode should be added inline to the output of the controller, unless of course one is already incorporated in the output stage of the controller to deal with this or other issues?

Just a theoretical experiment. Any ideas anyone? Chris, you seem to be the gooroo on this?

It has just occured to me that my thread on VFD questions has really morphed into DC motor theory, which would equally be relevant in this thread. Sorry Ewan, I didn't intend to hijack the thread!

Simon

Hi Simon,
I dont think you can parallel diodes like that. One get a little hotter than the others, resistance drops so it gets hot still until it fails, then the rest fail in a cascade. (will be interesting to see if I am correct lol)

Stuart

Simon: Stu's right you can't parallel diodes as you could get thermal runaway from uneven current sharing. I just used standard rectifier diodes as per those part numbers but there's nothing to stop you using a faster power Schottky diode. As far what's going on with the diode, the more I think about what's happening, the less sure I am about how it works. It seems to be a special case where throwing the switch means the diode is a short for the motor but not for the controller (as the diode wiring is fixed WRT to the controller, but not the motor), if that makes any sense. It's something I observed and also used without considering the underlying theory too much. The controllers this worked with had some form of voltage sensing feedback that the diode didn't affect.

I use it now (10A 1000V diode) on a simple PWM IGBT controller connected to a 2HP 180V treadmill motor for a 5 inch chop saw but then I'm not reversing it at full speed either. A quick test for your existing controllers to see if they have a back emf diode of some kind, is that with the controller switched off, the motor shaft will spin by hand easily one direction and have strong resistance the other way as the diode shorts the voltage generated by spinning the motor shaft.

OK guys I understand what you are saying in relation to paralleling diodes and the current not being exactly shared so one gets hotter than the other, reduces it's resistance and then more current, getter hotter again, rduces resistance...... What I don't follow is that even if the current is not shared it should not matter unless you are running 2 x 3A diodes for a 6 amp load where anything BUT a 50% share would cause failure. The main reason I don't understand is because it works with resistors does it not? I mean I have never had the need to do it with resistors but we were taught about
1/Rtot = 1/R1 + 1/R2

Where, even if you don't have the correct resistor value, you make make it up with combinations of other values. I'm now thinking that maybe it's only for small power situations like signal resistors and not for power dissipation applications?

Is my last thought correct then?

Graz, I understand that at the time you probably didn't put too much thought into how it worked and I'm not second guessing you or questioning your account of it, your explanation is not in question nor is the system you describe, I'm just trying to understand and get my head around how it works so that if/when I come to set one up, I don't let the magic smoke out of the controller the day I wire it up!

So, where do I get a 10 - 15A 1000V diode from? The more A's the better!

Cheers,

Simon

Hi Simon,

It depends...

What changes from diode to diode is the forward voltage drop, if one has a lower voltage drop it will take more current.

So paralleling 2 diodes won't necessarily double the current capacity. Might not increase the current capacity at all... It's easier to get a bigger diode.

I admit to not having read the whole thread, so apologies if this has already been discussed. Do you have a circuit of what you are trying to do?

Regards
Ray

EDIT: Here you go... Intersil RHR15120 15A 1200V Hyperfast diode 65nS | eBay

Hi guys, awake and feeling up to some thinking so here goes,

The next thing I was going to add to this thread was the braking circuit pic from the manual for the controller. It is simply a braking resister and switch (presumably if you want to turn braking off) and then a relay to turn the auto inhibit circuit in the board off whilst breaking is done. It is all very vague, the wiring diagram is poor, and there is not even a guide to resister sizes, but I found a site that has the maths on it and I think I only need a 100k 1.5w resister.

Simon, if you try eBay for a diode you may find one, or try rs online. Once you find a number searching is a lot easier. Often solar panels have big bypass diodes on them incase just that one panel is in shade or fails, that could be another thing to search for.

Mark, in your first post you say the diode acts as a brake, but after that you talking only about instant reversing. Being just a diode would be a hell of fast stop, the diode has to turn all that kenetic energy into heat quickly. If you wired a resister in series too though, that should slow the braking down and share the heat over both components and a longer time, meaning a smaller diode should be ok...?

R

Originally Posted by RayG

Hi Simon,

It depends...

What changes from diode to diode is the forward voltage drop, if one has a lower voltage drop it will take more current.

So paralleling 2 diodes won't necessarily double the current capacity. Might not increase the current capacity at all... It's easier to get a bigger diode.

I admit to not having read the whole thread, so apologies if this has already been discussed. Do you have a circuit of what you are trying to do?

Regards
Ray

EDIT: Here you go... Intersil RHR15120 15A 1200V Hyperfast diode 65nS | eBay

Hi Ray,
Mark pointed out using a diode across the dc out of the controller (against the flow, for the life of me all my anode cathode info has been sucked out of my brain) so that the controller can be reversed while the motor is still spinning without risk of frying the controller.

Hope you are ok Ray, sound like you trip was lot shorter than mine will be.

Originally Posted by Ueee

R

Hi Ray,
Mark pointed out using a diode across the dc out of the controller (against the flow, for the life of me all my anode cathode info has been sucked out of my brain) so that the controller can be reversed while the motor is still spinning without risk of frying the controller.

Hope you are ok Ray, sound like you trip was lot shorter than mine will be.

Hi Ewan,

That's a flywheel diode, what it does is absorb the back emf generated when the current through the motor switches off..

Most (all?) motor control fet's have flywheel diodes built in, and if it's a pwm type controller it should already have the flywheel diodes.

I hate hospitals... I went in Tuesday and got out on Sunday.

Regards
Ray

Originally Posted by Ueee

Hi guys, awake and feeling up to some thinking so here goes,

The next thing I was going to add to this thread was the braking circuit pic from the manual for the controller. It is simply a braking resister and switch (presumably if you want to turn braking off) and then a relay to turn the auto inhibit circuit in the board off whilst breaking is done. It is all very vague, the wiring diagram is poor, and there is not even a guide to resister sizes, but I found a site that has the maths on it and I think I only need a 100k 1.5w resister.

Simon, if you try eBay for a diode you may find one, or try rs online. Once you find a number searching is a lot easier. Often solar panels have big bypass diodes on them incase just that one panel is in shade or fails, that could be another thing to search for.

Mark, in your first post you say the diode acts as a brake, but after that you talking only about instant reversing. Being just a diode would be a hell of fast stop, the diode has to turn all that kenetic energy into heat quickly. If you wired a resister in series too though, that should slow the braking down and share the heat over both components and a longer time, meaning a smaller diode should be ok...?

Ueee, hope you're on the mend now. You could have a double pole three position switch with a suitable braking resistor on the middle position, that way going from forward to reverse has to go via the braking resistor in the middle. The diode would have to have a hard life conducting the stored motor inertia power but it allows a simpler/cheaper switch to be used.

Ray, a three day stay doesn't sound too good at all..... The original controllers were SCR based (they don't have internal diodes?) and later units were MOSFET. I have had a few MOSFETS die after the wrong voltage 400V external diode was fitted, I just suspected the diode died first (open circuit) and then the MOSFET went open circuit after it. Anyway it's only been a couple of units so far before the problem was fixed so I could only guess as to the failure mode. I just see the external diode as being a more rugged higher wattage support for the internal MOSFET diode.

Simon: It seems like a few experiments are in order to get an idea of whats happening, Uee's idea of a resistor in series with the diode is a good one, if a 240V 100W bulb was wired in series with the diode, it should not light in normal operation but if the reversing switch was used, then the bulb will glow if the diode is conducting and braking the motor??. I have to prototype a few boards of the new higher current controller board which is going to use IGBT's for the first time, I guess it's cheaper for me to smoke a few speed controllers in the name of science and prove it works for motors at significant power levels instead of the 1/15th HP motors I've used to date.