Lathe went down.

[ibuildstuff4u]

Just as I thought I was ready to use my new CNC lathe it shut down on me.  I was playing around with a G96 code which keeps the work piece at a constant surface speed by adjusting the spindle speed as the work piece changes size.  Everything was going good and then POP.  I blew a fuse.  I replaced it and as soon as you turn on the spindle it blows again.

I unplugged the motor from the speed control and it sill blew the fuse, so I unplugged the spindle controller and the fuse held.  It looks like the motor speed control is shot, but I wonder why it blew?  I took an ohm meter to the motor and there isn't any ohms to ground and the reading on the armature was around 6 ohms which sounds good, but I will have to see what it should read.  The only thing I wonder is if some how I blew it by playing around with the G96 code as it changes the 0 to 10 volt input quite rapidly as it adjusts the spindle speed.

The spindle speed control is a Minarik  PCM 2100A and is rated for a 1hp motor. 

If anyone has a suggestion as to why it could have blown I would love to hear it as I don't want to replace the controller and blow it out again.

Thanks, Dale P.

[John Rudd]

Dale,

You can check the motor if you have a variable dc power supply or a variac and bridge rectifier....Or some othe means of powering the motor with DC...A light dimmer perhaps.. If the motor runs ok, then it looks like the controller is broken..

You can test the controller by replacing the motor with a 100watt lamp(bulb or whatever you call it over there) If the bulb doesnt light as you adjust the speed or the fuse blows then the controller is broken ...
If you have access to a dvm and are familiar with electronics then it shouldnt take too much to determine where the fault lies with the controller.......

[DICKEYBIRD]

Dunno about your speed control/motor issue Dale but I'm pretty sure CSS is broken in Mach3.  I haven't tried it myself but everything I've read lately says it's broke and won't be fixed 'til Mach4.

Milton

[ibuildstuff4u]

Milton,

The CSS seemed to work for a while, but then for no reason it stopped working and just ran the spindle at full speed.  I closed Mach down and restarted it to try again and that's when the spindle started blowing fuses. 


John, 

I pulled the speed control and will check it out on the bench tonight.  There isn't to many parts to it so it should be pretty easy to figure out what died and perhaps just fix this one up, but it would be nice to have a smoking gun as to why it blew.  The CNC lathe sat for several years un used when I bought it so I may pull the motor out and take a good look at it too. 

I like your idea of putting together a make shift power supply to test the motor and will probably give that a shot as I should have all the parts I need right here in the shop.

Thanks for the help!
Dale P.

[PekkaNF]

.... The CNC lathe sat for several years un used when I bought it ....
Dale P.

If the speed controller hadn't been used on on long time and then fails I would check first electrolyte capcitors, specially on the PSU section. Any buldging or anyt other sign and they could be shot. If you are lucky, they might show shortted, but often they just are somewhat bad and then fail when under real conditions. I had one SMPSU that refused to work and Multimeter passed all capacitors fine. It did start to work when I changed all secondary side capacitors, least one of them was bad I gues.

This is just a tough, yours might be something else, but capacitors are usually easy to find and replace.

Pekka

[ibuildstuff4u]

Capacitors are a good idea, but there really isn't any in this power supply other than a few really small ones. 

I bench tested the spindle drive and I can't even power it up with out blowing a breaker.  This is with NOTHING plugged into the drive but power in.  I checked out the SRC's and diodes and to me it looks like I have two blown diodes.  The question is WHY.  What happened, and how do I prevent this from happening again.

I pulled the motor from the machine to inspect and found it to be in great condition.  The brushes are hardly worn and armature looks great.  The pulley looks like it was rubbing a bit which helps nail down a noise I heard when the machine was first turned on for the day and cold, but other wise the motor and wires look fine.

I looked into some info on the web and found this in the instructions.  "Do not change jumper settings with power applied. Failure to follow these instructions may result in damage to the motor or drive, or serious injury to personnel."

This jumper that they refer to changes the control from a pot switch to a 0 to 10 volt control so the computer can operate the spindle speed.  While I was using the lathe I had moved the dial from the computer control to the pot switch control a few times before it went down and must have caused the problem.  I'm surprised that Prolight has this dial set up so a user can change the setting with out any warnings that it can cause damage if done so while powered up.  I even have the users manual and it doesn't say anything about it causing an issue.

Hopefully this was the cause of the drive failure and this won't happen again as it's not only expensive to replace, but I don't need the down time.

Dale P.

 

[John Rudd]

Determining root cause failure of electronic components isnt easy....

In your case though I'd ask which  two diodes have failed?
In my experience, dc speed controllers usually have two scr's and 2 diodes arranged in a full wave bridge configuration and if the design is such that it has a seperate excitation supply for a field winding, then 2 seperate diodes are used for that which are generally of a lower current rating the the diodes in the bridge...However the PIV rating may be too low, a spike on the supply line may have caused the damaged...Solution....use two diodes of a higher PIV rating and if not already fitted, add a transient suppressor across the mains input with a suitable voltage rating..

[DICKEYBIRD]

Dale, here's a thread on Mach & CSS you may find interesting.  It confused me back when I read it originally and still does.  I have a LONG way to go with this stuff.

Fortunately the work I'll be doing is small & mostly aluminum so CSS isn't a critical thing for me.

[ibuildstuff4u]

John,

Here is a picture of the controller taken off it's heat sink.  There are two SCR's and three diodes in the middle of them. (I'm pretty sure they are diodes as they are labeled D506, D507 and D508 and only have two wires, but looking at it closer I see they use to have three wires and one wire was clipped off.) They say 071-0039 on them and a D2, but that doesn't seem to reference anything when I try to google it.  I'm guessing Miniark has their own label on them so you have to go to them for replacements.  The two outer "diodes" read shorted on a meter, and the middle one tests as a diode should, no reading one way and around 650 ohms the other direction.  Having the two outer ones read a dead short makes me think they are at fault.  They are right inline where the power enters the board too.

On the back I see four diodes in a bridge rectifier formation, but they seem too small for the motor and must be for the signal transformer to convert AC to DC for the circuit on back.  I tested all of the diodes and they all seem fine as well as the two small diodes next to the input transformer on the top side of the board. 

I'm going to order a new controller this week, but would also like to repair this one for a spare.  Hopefully I can find some replacement parts for a reasonable price.

I would be very interested to know how to add a transient suppressor to the main input as sometimes the motor kicks in hard when first started and it blows my GFI outlet.


Milton, I will take a look into the reading you suggested.  I also work with aluminum 99% of the time, but thought it would be cool to use the G96 code and run like the big machines.  I have been reading the Mach 3 manual a lot lately and see there are a ton of features that would be neat to use.  Hopefully I can figure out how to get some of them to work!

Dale P.

[ibuildstuff4u]

After lots of clicking, I found a link that says the blown parts are Minarik's own part # and are a 800 V, 20 A Diode.  I will have to look and see if I can find one in a 220 case.

Dale P.

[trevoratxtal]

""looks like I have two blown diodes"".
I would concur with John Rudd
Diodes are often poorly rated for the job.
If you rapidly change voltage on a inductive load the back EMF spikes can be very large 10 or even 100 times the normal run voltage.
Replace the diodes  with the highest voltage in that type. (cost is negligible)
If you give the number on the diode I can look up a better replacement.
Also a small capacity ie 0.01 µf, high voltage (1000volt non electrolytic ) capacitor is wired across the brush contacts it will reduce the damaging spikes.
Good luck
Trev

[Bluechip]

800 V, 20 A Diode.  I will have to look and see if I can find one in a 220 case.

Dale P.

Littelfuse D8020L IIRC
 
Comme ca:
 
http://uk.farnell.com/littelfuse/d8020l/diode-standard-20a-800v-to-220/dp/1456539?ref=lookahead
 
Despite the piccy, it does stand on two legs ..    er ... mostly.
 
Dave BC
 
 

[trevoratxtal]

Can you read the numbers on them?
Trev

[ibuildstuff4u]

The only numbers on the part is 072-0039  which according to my research this is Miniark's own part #.

[DICKEYBIRD]

Depending on how much another Minarik costs, you may want to switch to a KB controller.  They seem to be (almost) bulletproof with very good online manuals.  I have 3 KBIC-120's that are rated at 1 HP and they've worked well with everything I've thrown at them.  I got all of them used on ebay for $25 to $35 each.

The speed control pot circuit isn't isolated (I think it is on the Minarik) so you would have to deal with that but you being electrically knowledgeable wouldn't have any trouble.  That's why I said (almost) bulletproof.  Ignant souls like me can kill anything!

[ibuildstuff4u]

I replaced the controller on my X2 milling machine with the KBIC-120 and it works way better than the original.  I get a lot more torque and RPM out of the motor now.  I will look into the KB drives, but it will have to have an input for a 0 to 10 volt control as I really like the machine being able to run the spindle speed.  I'm not sure if the 120 has that, but I will take a look.

Dale P.

[DICKEYBIRD]

Yep, definitely has voltage following capability but no isolation.  I had my -120 working well from Mach within 20 to 50 rpm across the range.  I've switched to another KB with reversing module and dynamic braking now.

What controls the Minarik; your B.O.B?  I forget what you have in yours.

[ibuildstuff4u]

I went with a breakout board from Sound logic.  It's their combo board which has four relays built in as well as the 0 to 10 volt control for the spindle speed.  The breakout board is kinda pricey at $250.00, but it had every thing I needed as well as built in power supplies to power the breakout board and an extra 12 volt source that can be used for extra relays and proximity switches. You don't need two different 5 volt power supplies with this breakout board to keep the computer isolated because it has them already built in.  Just power it up with 110 VAC and your good to go.

There is a large external resistor hooked up to my spindle speed control to provide breaking so I already have that, and as far as reverse goes the chuck is threaded on with no lock so I can't do it.  I guess I could make some type of set screw to prevent it from backing off, but I don't see the need since I don't have a fancy second tool post like yours.

Isn't your new drive for AC motors too?  I thought I read that in the spects when you got it which means it wouldn't work with mine since I have a DC motor. 

[ibuildstuff4u]

Dave I took a look at your link for the Diode and it looks like a good fit to me!  I looked at some other ones on the net but they used different legs which would have been harder to replace.  I see they make a 1000 volt 25 amp model which would step up the ratings a bit.

I will have to see if mouser or Digi-key has them in stock as your source is on the other side of the pond.  I will pick up 6 so I can replace all three and have three left for spares.

Thanks guys for all the help!
Dale P.

[andyf]

"Newark element 14" in Chicago is part of the Farnell Group, and stock the D8020L. The 8 indicates 800V PIV, the first 0 doesn't seem to mean anything and the 20 at the end indicates 20A.
< http://www.newark.com/littelfuse/d8020l/diode-standard-12-7a-800v-to-220/dp/11J8453?ref=lookahead >
They do stock a 1000V version called the DK025L
< http://www.newark.com/littelfuse/dk025l/diode-standard-15-9a-1kv-to-220/dp/11J8352?Ntt=DK025L >
but it's only rated for 15.9A.

The D8020L is used in the higher rated KB boards like the KBIC 240 I put on my mill, according to an old KB circuit diagram I have.

Finally, they do have three legs, but if you imagine the ones shown in the above links standing upright, the RH leg isn't connected to anything and just seems to provide extra suppost.

Andy

[ibuildstuff4u]

Andy, looking at the data sheet it says

The D8020L has a IF(AV) rating of 12.9 amps and a IF RMS rating of 20 amps.
The DK025L has a IF(AV) rating of 15.9 amps and a IF RMS rating of 25 amps.
D for who knows what, K for thousand volts and 25 for amps?

I'm not sure what IF(AV) and IF RMS stands for but to me the DK025L is higher. Everything else seems to be the same so the larger DK025L should work?   I also saw that it comes with the third leg, and Minarik simply sniped it off and doesn't offer a hole for it to solder in.  Some of the other diodes I looked at only had two legs, but placed them in a different configuration which wouldn't have fit into the board with out some funky bends in the leads so I'm glad this one was pointed out to me.

I'm just about to pull the trigger on ordering some DK025L's from Mouser as they have them in stock, but I will wait a bit longer to see if you have an opinion on going with the DK025L's.

Dale P.

I thought the lathe had a pretty nice looking motor though!  It's a 1HP Baldor with an encoder attached to run the tach on the front of the machine.

[DICKEYBIRD]

Isn't your new drive for AC motors too?  I thought I read that in the spects when you got it which means it wouldn't work with mine since I have a DC motor.

Mine definitely is for DC motors, PM or wound fields.  I have a Baldor 3/4 hp, 1750 rpm DC PM motor.

Looking at the picture of yours with the reduction belt drive and smaller diameter motor, I'm guessing yours is a higher speed Baldor...3500 rpm?  Nice!

[ibuildstuff4u]

Wouldn't you know I just put the motor back in and never read the plate on it. 

So I don't know how fast the motor spins, but I do know the spindle will do over 1200 RPM in low speed and 3600 RPM in hi range.  More than fast enough for me.  I think I'm running my manual lathe at 650 RPM and the finish is always nice. 

[Bluechip]

Andy, looking at the data sheet it says

The D8020L has a IF(AV) rating of 12.9 amps and a IF RMS rating of 20 amps.
The DK025L has a IF(AV) rating of 15.9 amps and a IF RMS rating of 25 amps.
D for who knows what, K for thousand volts and 25 for amps?

I'm not sure what IF(AV) and IF RMS stands for but to me the DK025L is higher. Everything else seems to be the same so the larger DK025L should work?   I also saw that it comes with the third leg, and Minarik simply sniped it off and doesn't offer a hole for it to solder in.  Some of the other diodes I looked at only had two legs, but placed them in a different configuration which wouldn't have fit into the board with out some funky bends in the leads so I'm glad this one was pointed out to me.

I'm just about to pull the trigger on ordering some DK025L's from Mouser as they have them in stock, but I will wait a bit longer to see if you have an opinion on going with the DK025L's.

Dale P.

I thought the lathe had a pretty nice looking motor though!  It's a 1HP Baldor with an encoder attached to run the tach on the front of the machine.

 
'I' is the SI dimension for current , the 'F' is forward.
 
More usually done with subscript thus ...  I_{FWD
 
R.M.S is just 'root-mean-squared'
 
The difference is due to 'Form Factor'   says he, valiantly trying to remember stuff he did half a century ago ...
 
Dave BC}
 
 

[andyf]

Dale, it usually does no harm to substitute with a component which is overated in terms of the volts, amps or watts it will take, if its other parameters are the same.  A simple example would be replacing a 0.5 watt, 10 ohm resistor with a 10 watt, 10 ohm one. Using the DK025 rather than the D8020 will just mean that the component is less likely to approach the danger zone, so it should last longer. The D seems to stand for Diode; the same manufacturer make a complementary range of SCRs and prefixes those with S as in S8020. The SCRs on your boad may be those.

Fuses shouldn't be overrated, of course; replacing a 1A one with a 10A could/would be hazardous.

Point taken over the current rating. I think the lower values (12.9A and 15.9A) are what the components would take if you put a pure DC current through them. They will take a higher current if if they are used to rectify AC, because they are passing no current at all for half the time and so won't get as hot as they would with pure DC.

I'm getting outside my rather sketchy knowledge of the subject here, but I think the RMS (root mean square) value relates to AC currents with a sinusoidal wave form where current and voltage are in phase, like in your domestic  mains supply. The maximum current would be less if the waveform was square.

Andy

[Bluechip]

Doesn't have to be sinusoidal Andy ... you can get the r.m.s. of anything repetitive if you want to .
http://en.wikipedia.org/wiki/Root_mean_square
You don't have to stick with the ' root two ' bit .
All that stuff meant some thing to me once. I was OK with electrical stuff once, but it all changed after Faraday poked his nose in then I lost the plot somewhat ... 
 
Dave BC
 

[DICKEYBIRD]

...with an encoder attached to run the tach on the front of the machine.

Hmmm, I wonder what the encoder out put is?  Some of the KB's (incl. my KBCC, yay!) accept a 7VDC or 50VDC per 1000 rpm tach input and will regulate the speed to <1% of set speed.  If I have trouble with speed wobblies when threading, I'll hook up one of my old PM brushed ball-bearing model airplane motors to the Baldor & use it as a tach generator.

[andyf]

..... All that stuff meant some thing to me once. I was OK with electrical stuff once, but it all changed after Faraday poked his nose in then I lost the plot somewhat ... 
 
Dave BC

Me too, Dave. Had to learn all that sort of stuff for the Amateur Radio exams, but that's about 30 years ago. I got distinctions, but that's nothing to boast about; the questions were tick the box multiple choice, and included things like this:

Q1.
[schematic of something]
What does this diagram show?
(a) a common base amplifier (b) a common collector amplifier (c) a low pass filter (d) a power supply

Q2
In the diagram of a common collector amplifier shown in Q1, what is the function of R1?
(a)        (b)          (c)            (d)

Andy

[John Rudd]

When sourcing replacement diodes or scr's for that matter, remember the KB or the Minarik decies are isolated tab....  Dont think of fitting anything else without using isolation bushes and micas if you need to else you will continue to blow fuses

[ibuildstuff4u]

 John I wondered about the isolation on the tab because there isn't any type of isolation on this drive and I'm pretty sure I'm the first to take it apart.  There is just a thin film of thermal compound between the tab on the diodes and scr's to the heat sink.  There are NO bushings, or micas between them or any evidence that their was ever one there as the micas usually leave a mark larger than the tab on the heat sink when you remove them.

Sounds like I should add some isolation?

Dale P.



Milton,  I thought it was funny that the lathe had an encoder for a built in Tach and also a optical pickup to provide the computer with the spindle speed.  Seems like there would be an easy way to provide the computer and RPM gauge with the same signal.  It's also strange as both gauges aren't the same and are off by about 20 to 30 RPM, but that could be due to the pulley ratio on the encoder verses the optical pickup being mounted right on the spindle.

It sure would be neat to have the speed controller hooked up to the encoder so the speed is more constant to what you set it to.  I still have to spend time on my machine to get the RPM to match what the computer is asking it to be.

Dale P.

[John Rudd]

I wondered about the isolation on the tab because there isn't any type of isolation just a thin film of thermal compound between the tab on the diodes and scr's to the heat sink.  There are NO bushings, or micas between them
Dale P.

Dale,

Perhaps I wasnt as concise as I thought...
The OEM devices have an isolated tab....Nothing else required....

If you obtain devise from another manufacturer, they may need micas and bushes..

[ibuildstuff4u]

Ok I get it.  Makes sense now.  Thanks for the tip, I will check out the info sheet on the diodes and see if the tab is isolated or not.  Guess I could just test them with a meter too, or just insulate them to be on the safe side.

Thanks for the tip, Dale P.

[trevoratxtal]

ibuildstuff4u I dropped out of this post As John and others were giving good advice particularly to supply's your side of the pond.
It is very gratifying to see so many Electrical/Electronic knowledgeable folk.
I salute you all. 
Trev
PS do not forget a capacitor across the brushes on DC motors. It also cuts down RFI.

[ibuildstuff4u]

Your guys rock!  I didn't know a lot of you guys are interested in electronics too!  I loved playing around with the stuff as a kid and am interested in getting back into electronics as well as doing more metal working in the future.  I think I'm going to add a lab area to my shop for electronics and pick up some equipment such as a nice variable power supply and an oscilloscope. 

I'm 37 years old and have been going to night school for the past few years to be a machinist, but recently changed my degree to an Automated Systems Technician and I can't wait to take some of the core classes and learn more about machine controls, and electronics in general.  Right now I'm working my way through a math class that I take Saturday mornings and it has been tough!  It's all self paced with NO formal instruction so you have to teach your self and work on homework 3 to 4 days a week. 

Trevoatxtal,   When I add the caps, do they have to be right by the motor, or can I place them closer to the speed control?  Also do they get mounted one from neutral to ground and then a second one mounted from the hot to ground?  Or do the go right across the hot and neutral leads?  Years ago I ran RC cars and we added caps to the motor to prevent interference to the radio and the caps were soldered right to the motor case.  One cap went from the + lead to the motor case and the second cap went to the - lead to the motor case.

Thanks again for all the help!  My new parts come on Wednesday so hopefully that's all I need to get it working again.

[John Rudd]

If the motor is earthed( grounded) then connect a cap from each motor terminal to the casing....

I hope the caps you are using are not polarised......A suitable value would be around 100nF (0.1 mfd_

[trevoratxtal]

I concur with John Rudd but add must be high voltage working 1000v min, you could get away with .01µf  with the frequency's the controller runs at but .1µf is better but 10 times bigger physically so harder to place.
Good luck
Trev  

[ibuildstuff4u]

The parts came today and the lathe is back up and running again!  It's always a great feeling when you can fix something instead of having to buy a new one.  It put a nice big smile on my face when I pushed the button to start the spindle and the lights didn't go out. 

Thanks again for all the help!

Dale P.

[andyf]

  We like stories with happy endings!

Andy

[ibuildstuff4u]

I just bumped in to this on E-bay.  For around $100.00 I can get the whole box that contains the motor controller.  It would provide me with a spare motor drive as well as a, RPM gauge, and two solid state relays.

Very tempting, but funds are tight right now.  I will have to sleep on it.

Dale P.

http://www.ebay.com/itm/261197635324?ssPageName=STRK:MESINDXX:IT&_trksid=p3984.m1436.l2649