• This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn more.
  • PLEASE: Read the FORUM RULES BEFORE registering!

4

Magnetic Chuck Control Circuit Wanted

3
Like what you see?
Click here to donate to this forum and upgrade your account!
10

markba633csi

Active Member
Active Member
Joined
Apr 30, 2015
Messages
1,317
Likes
526
#61
The PTC (positive temperature coefficient) device is the same little gizmo that they used to control the degaussing coils around your color tv picture tube. It tapers the current down as it heats up. John H. and I (and others) were thinking it would give a better demag effect than just a shot of AC because if your shot happened to stop right on the peak of the AC sinewave- well you get the idea. Of course if you have a variac in the circuit you could just flip the switch to demag and turn the voltage down. Same effect.
The MOV (metal oxide varistor) is a voltage clamp like a double zener diode. Protection for the bridge rectifier and toggle sw. from inductive kick from the chuck coil. Small 0.1uf cap also. Omit them if you dare LOL Someone else (Pete?) zapped a bridge already. But his setup was somewhat unusual.
Extra filtering caps- optional as far as I'm concerned. Another part to dry out and fail someday. If you can get good results without it...Large high voltage electrolytics are always the weak link- and expensive. Use the smallest uf value you can.
Mark
 
Last edited:

markba633csi

Active Member
Active Member
Joined
Apr 30, 2015
Messages
1,317
Likes
526
#62
Be careful out there everyone- don't do the 60 cycle shuffle unless you have a partner.
MS
 

Bob Korves

H-M Supporter - Premium Member
H-M Supporter-Premium Member
Joined
Jul 2, 2014
Messages
4,032
Likes
4,182
#63
I was studying this while you were writing, Mark. OK, I understand using the variac and dialing it down to demag, bypassing the rectifier. I like that idea, and have the stuff to do it.

I thought that a fairly large cap was needed to smooth the DC ripple created by the rectifier. It sounds like I may be missing something here... My electronic knowledge leaves a lot to be desired.
 

Rick Berk

Active User
Active Member
Joined
Feb 5, 2013
Messages
145
Likes
81
#64
YES, I'm still here and learning, think I have located a variac, will know the details by Saturday.
 

whitmore

Active Member
Active Member
Joined
Mar 3, 2017
Messages
173
Likes
92
#65
I thought that a fairly large cap was needed to smooth the DC ripple created by the rectifier.
A smoothing of the DC ripple is also accomplished by use of an inductor, like... the
hulking electromagnet. It is so effective at smoothing current, in fact, that
the switch and Variac windings are likely to get arc discharge damage, if the MOV is
omitted from the circuit. A large capacitor would smooth low-frequency ripple,
but a small one prevents RF switching hash (so the Variac adjustment doesn't
put static onto every AM radio in the vicinity). An early schematic has small
capacitors across every diode in the bridge rectifier (a sure sign that someone
is a ham, IMHO).
 

markba633csi

Active Member
Active Member
Joined
Apr 30, 2015
Messages
1,317
Likes
526
#66
:D No, never keyed up, actually. A little shortwave listening only. One can only have so many hobbies LOL
I think whitmore and I are on the same frequency. I didn't know if a big cap would be needed on the dc output so I didn't show one.
If you do use a big filter cap be sure to hang it across the bridge rectifier output not the chuck terminals. You don't want to put AC on it. AC is ok for the little cap and the MOV.
MS
 
Last edited:

whitmore

Active Member
Active Member
Joined
Mar 3, 2017
Messages
173
Likes
92
#67
:D I didn't know if a big cap would be needed on the dc output so I didn't show one.
If you do use a big filter cap...
The magnet, as inductor, takes the rectified AC and makes its RMS value
(about 120V) average across the load. A big filter cap, takes the rectified AC
and potentially makes the peak (170V) appear, on average, across the load. That
exceeds the expected applied DC voltage... by too much. The filter capacitor
is not particularly useful, unless the AC ripple on the inductor causes
hum (in some way that disturbs the cut). If that were to be an issue,
a stepdown transformer, or variac with a stop, would fix it, at some cost.

The switching of a 170VDC power source is hard to accomplish: most
switches cannot handle DC voltages that high. The DigiKey offerings top out
at 125 VDC for toggle switches (but 277VAC if switching alternating current).
 

MozamPete

Active User
H-M Supporter-Premium Member
Joined
Jun 26, 2014
Messages
500
Likes
800
#68
The magnet, as inductor, takes the rectified AC and makes its RMS value
(about 120V) average across the load. A big filter cap, takes the rectified AC
and potentially makes the peak (170V) appear, on average, across the load. That
exceeds the expected applied DC voltage... by too much.
The inductance of the magnetic chuck will tend to smooth out and average the current flowing, not the voltage. Without a capacitor the voltage waveform will still be the same raw rectified AC voltage and you will still be exposing the coils to the 170V peak, just only for a short period 120 times a second (instead of continually if you had smoothing capacitor).

This has been my main problem with trying to develop a transformerless design - and exasperated by the fact I'm starting with a 230V ac supply over here so have a potential 325V peak.
 

whitmore

Active Member
Active Member
Joined
Mar 3, 2017
Messages
173
Likes
92
#70
The inductance of the magnetic chuck will tend to smooth out and average the current flowing, not the voltage. Without a capacitor the voltage waveform will still be the same raw rectified AC voltage and you will still be exposing the coils to the 170V peak, just only for a short period 120 times a second (instead of continually if you had smoothing capacitor).
Yes, the current (which determines resistive heating) is limited by the 120V average
if there's no big filter capacitor. Resistive heating, and overcurrent causing
saturation effects, are things to avoid. Overvoltage, on the other hand... that's
no problem for any reasonable wiring insulation: all the wiring in motors and
such gets kilovolts applied for safety testing. The peak DC voltage doesn't
threaten the electromagnet. The average voltage, though, does.

A variant on a light dimmer (the kind of light dimmer that runs transformer-operated
lights) can effectively lower average voltage, by switching the AC input. Another way
to feed less voltage from a too-high-voltage input, is to use a ballast (like,
the old-fashioned magnetic fluorescent ballasts) before the rectifier.
One might build a decent 100V/1A current source by connecting AC through
old-style magnetic fluorescent ballasts (one or more in parallel).

Dimmers are, alas, likely to fail in full-brightness mode, so fuses are not
optional if the dimmer is the only voltage limit.
 

Bob Korves

H-M Supporter - Premium Member
H-M Supporter-Premium Member
Joined
Jul 2, 2014
Messages
4,032
Likes
4,182
#71
I already have, on hand, two variacs.

The first one shows either:
Input: 120V 50/60hz -- Output: 0-120v -- Amps 2.25 -- V.A. 270
or
Input: 120V 60hz -- Output 0-132V -- Amps 2.25 -- V.A. 297

The second one:
Input 120V 50/60hz
Output: 0-140V KVA =1 (big mutha')

I also have a 450V, 470 μF electrolytic cap, 105C

I also have a drum switch and a more than capable rectifier bridge, and a bleed resistor...

Whitmore and Mark, tell me where I might go next with this start...
 

whitmore

Active Member
Active Member
Joined
Mar 3, 2017
Messages
173
Likes
92
#73
I already have, on hand, two variacs.

The first one ...
Input: 120V 60hz -- Output 0-132V -- Amps 2.25 -- V.A. 297

The second one:
Input 120V 50/60hz
Output: 0-140V KVA =1 (big mutha')

I also have a 450V, 470 μF electrolytic cap, 105C
...a drum switch and a more than capable rectifier bridge...
Well, MozamPete says the 8x24 incher (192 square inches) takes 130W,
so your 5x11 (55 square inches) ought to be comfortable with 38W.
To start, get a fuse (1/2A, maybe less) and run the Variac output
through the fuse into the rectifier. Check that the chuck terminals (there
should be two, and a ground) don't have a short to ground, then
connect the rectifier output to (1) a DC-voltmeter and (2) through a
DC-ammeter, into the chuck. MOV at the chuck, if you have one.

Start at something safe (20V?) on the variac, and note current and voltage readings,
preferably on graph paper... take small steps (2 to 5V) and plot a few V-versus-I points,
as you turn up the Variac.

If the chuck starts to buzz, it's probably magnetically saturating (and further
current increase is counterproductive). Note the variac reading, and DC voltage.

If you get all the way to 120 on the Variac and the fuse hasn't blown, shut down.

Now, look at the graph: if it's just a straight line, no problem. If it starts
straight, and curves UP to higher current at some voltage, note the onset
of that curve and subtract 10 percent. The upcurve happens when the magnet
starts to saturate, you do NOT want that to happen in operation. It's
equivalent to overloading a motor, which hums, heats, emits smoke...
The ripple current after the rectifier will upswing sharply at saturation.

Speaking of which, even if the fuse doesn't blow, if you smell something
from the magnetic chuck, shut down. Obviously.

The idea here, is just to get a current and voltage operating point.
It would be great if you could put an AC ammeter in series with the DC
ammeter: the AC ammeter will show a strong indication when magnetic
saturation occurs, and the DC ammeter will tell you the operating current to stay well under.
The AC ammeter will also jump every time you adjust the Variac.
 
Last edited:

Bob Korves

H-M Supporter - Premium Member
H-M Supporter-Premium Member
Joined
Jul 2, 2014
Messages
4,032
Likes
4,182
#74
Bob what are the specs again on your mag chuck? (Input voltage and amps or watts)
MS
No tags. Resistance is 86 ohms across the the plug on the electrical cable. Everything is properly soldered in the chuck, so that number should be usable. I am sure it was meant to be fed 120VDC. I am guessing(!) that I would not want to run more than 120 watts (1 amp) to the chuck at 120VDC. The variacs will give me some wiggle room there...
 

Bob Korves

H-M Supporter - Premium Member
H-M Supporter-Premium Member
Joined
Jul 2, 2014
Messages
4,032
Likes
4,182
#75
Well, MozamPete says the 8x24 incher (192 square inches) takes 130W,
so your 5x11 (55 square inches) ought to be comfortable with 38W.
To start, get a fuse (1/2A, maybe less) and run the Variac output
through the fuse into the rectifier. Check that the chuck terminals (there
should be two, and a ground) don't have a short to ground, then
connect the rectifier output to (1) a DC-voltmeter and (2) through a
DC-ammeter, into the chuck. MOV at the chuck, if you have one.

Start at something safe (20V?) on the variac, and note current and voltage readings,
preferably on graph paper... take small steps (2 to 5V) and plot a few V-versus-I points,
as you turn up the Variac.

If the chuck starts to buzz, it's probably magnetically saturating (and further
current increase is counterproductive). Note the variac reading, and DC voltage.

If you get all the way to 120 on the Variac and the fuse hasn't blown, shut down.

Now, look at the graph: if it's just a straight line, no problem. If it starts
straight, and curves UP to higher current at some voltage, note the onset
of that curve and subtract 10 percent. The upcurve happens when the magnet
starts to saturate, you do NOT want that to happen in operation. It's
equivalent to overloading a motor, which hums, heats, emits smoke...
The ripple current after the rectifier will upswing sharply at saturation.

Speaking of which, even if the fuse doesn't blow, if you smell something
from the magnetic chuck, shut down. Obviously.

The idea here, is just to get a current and voltage operating point.
It would be great if you could put an AC ammeter in series with the DC
ammeter: the AC ammeter will show a strong indication when magnetic
saturation occurs, and the DC ammeter will tell you the operating current to stay well under.
The AC ammeter will also jump every time you adjust the Variac.
Thanks, whitmore! That sounds like a very good way to test the chuck for how much power it can accommodate before it saturates. I will do that and report back, may take a while to find enough components to make a reasonably safe and worthwhile test run. The chuck body has no connection to the power wires, already tested...
 

markba633csi

Active Member
Active Member
Joined
Apr 30, 2015
Messages
1,317
Likes
526
#76
Sounds good. You don't HAVE to have an isolated supply to run your chuck, but it is safer from a shock hazard standpoint. Code would require it. But in your own
house you can decide. Just make sure the chuck is well grounded whichever way you go.
Cheers,
MS
 

Bob Korves

H-M Supporter - Premium Member
H-M Supporter-Premium Member
Joined
Jul 2, 2014
Messages
4,032
Likes
4,182
#77
Sounds good. You don't HAVE to have an isolated supply to run your chuck, but it is safer from a shock hazard standpoint. Code would require it. But in your own
house you can decide. Just make sure the chuck is well grounded whichever way you go.
Cheers,
MS
I plan to change the cord to a three wire one, and ground to the chuck casing. The existing cord is at least 50 years old and is petrified anyway, but interestingly has no cracks or other issues, just very stiff. To hell with code, I do it to standards that are considered safe by those in the know... It is my tail that is on the line! Thanks!
 

Bob Korves

H-M Supporter - Premium Member
H-M Supporter-Premium Member
Joined
Jul 2, 2014
Messages
4,032
Likes
4,182
#78
Well, MozamPete says the 8x24 incher (192 square inches) takes 130W,
so your 5x11 (55 square inches) ought to be comfortable with 38W.
6x18" chucks typically draw about 100-120 watts, maximum.
 

MozamPete

Active User
H-M Supporter-Premium Member
Joined
Jun 26, 2014
Messages
500
Likes
800
#79
No tags. Resistance is 86 ohms across the the plug on the electrical cable. Everything is properly soldered in the chuck, so that number should be usable. I am sure it was meant to be fed 120VDC. I am guessing(!) that I would not want to run more than 120 watts (1 amp) to the chuck at 120VDC. The variacs will give me some wiggle room there...
The chuck operating power should pretty much just be based on the DC resistance of the winding - in steady state operation on dc the inductance of the winding has no effect. So

I = V/R and
power = V²/R

At 120Vdc and 86 ohms take would imply

Current = 1.4 Amps
Power = 167 Watts


For comparison my 8" x 24" chuck name plate is 110Vdc, 1.3A, 144W and measuring the dc resistance of the chuck I get 92 Ohms.
 

Bob Korves

H-M Supporter - Premium Member
H-M Supporter-Premium Member
Joined
Jul 2, 2014
Messages
4,032
Likes
4,182
#80
The chuck operating power should pretty much just be based on the DC resistance of the winding - in steady state operation on dc the inductance of the winding has no effect. So

I = V/R and
power = V²/R

At 120Vdc and 86 ohms take would imply

Current = 1.4 Amps
Power = 167 Watts


For comparison my 8" x 24" chuck name plate is 110Vdc, 1.3A, 144W and measuring the dc resistance of the chuck I get 92 Ohms.
Yes, and that sounds like too much power to me. Perhaps the voltage needs to be lower. The variac can do that easily. I like whitmore's method of testing for saturation, and I will do that.
 

hman

Active User
H-M Supporter-Premium Member
Joined
Feb 17, 2013
Messages
1,530
Likes
1,045
#82
Hello to you, and welcome! Do you have a surface grinder?
 

awright

Swarf
Registered Member
Joined
Jul 20, 2016
Messages
4
Likes
0
#84
Well, MozamPete says the 8x24 incher (192 square inches) takes 130W,
so your 5x11 (55 square inches) ought to be comfortable with 38W.
To start, get a fuse (1/2A, maybe less) and run the Variac output
through the fuse into the rectifier. Check that the chuck terminals (there
should be two, and a ground) don't have a short to ground, then
connect the rectifier output to (1) a DC-voltmeter and (2) through a
DC-ammeter, into the chuck. MOV at the chuck, if you have one.

Start at something safe (20V?) on the variac, and note current and voltage readings,
preferably on graph paper... take small steps (2 to 5V) and plot a few V-versus-I points,
as you turn up the Variac.

If the chuck starts to buzz, it's probably magnetically saturating (and further
current increase is counterproductive). Note the variac reading, and DC voltage.

If you get all the way to 120 on the Variac and the fuse hasn't blown, shut down.

Now, look at the graph: if it's just a straight line, no problem. If it starts
straight, and curves UP to higher current at some voltage, note the onset
of that curve and subtract 10 percent. The upcurve happens when the magnet
starts to saturate, you do NOT want that to happen in operation. It's
equivalent to overloading a motor, which hums, heats, emits smoke...
The ripple current after the rectifier will upswing sharply at saturation.

Speaking of which, even if the fuse doesn't blow, if you smell something
from the magnetic chuck, shut down. Obviously.

The idea here, is just to get a current and voltage operating point.
It would be great if you could put an AC ammeter in series with the DC
ammeter: the AC ammeter will show a strong indication when magnetic
saturation occurs, and the DC ammeter will tell you the operating current to stay well under.
The AC ammeter will also jump every time you adjust the Variac.
 

awright

Swarf
Registered Member
Joined
Jul 20, 2016
Messages
4
Likes
0
#85
Whitmore, I'm not sure that you will get the desired indication of saturation in your graph if you are relying only on a DC ammeter. Granted, at saturation the incremental inductance disappears (or is reduced immediately or gradually, depending upon the hysteresis curve of the magnetic core), but because the DC current is determined only by the applied voltage and the DC resistance of the wire in the coil, not by the inductance, I would think the DC I/V curve would be immune to magnetic saturation effects. There would, however, be an indication on an AC ammeter looking at the ripple current if you are using an unfiltered, simple rectifier power supply. Having a sensitive tong type AC ammeter on the coil wire would give a useful indication of saturation, assuming an unfiltered supply. AC tong tools are relatively cheap these days.

You do mention the desirability of having an AC ammeter in the circuit and the fact that it will indicate saturation, but I think an AC ammeter, not the DC I/V plot, should used as the primary indicator of magnetic saturation.

However, I think the thermal dissipation of the coil would be the primary limiting factor on permissible applied DC voltage, probably long before saturation in a properly designed chuck. This is fairly difficult to evaluate non-destructively unless you have temperature sensor imbedded within the coil windings, something that you can only accomplish during winding of the coil. There would be an upward bend in the DC I/V curve due to heating of the wire, but thermal damage might occur before you saw the effect. One might get a useful indication of maximum permissible current by inserting a temperature sensor as deeply into the windings as possible (probably being only at the surface, depending upon the tidiness of the windings and the presence of any potting resin), and VERY GRADUALLY increasing the voltage/current while watching the temperature and feeling the surface of the chuck. By gradually, I mean over many hours, as the temperature will be maximum in the middle of the windings and it will take time for the surface to indicate possibly limiting temperature in the core.

Be aware that I am not an electrical engineer (or even a magnetic or thermal engineer - or even a professional machinist).
 

awright

Swarf
Registered Member
Joined
Jul 20, 2016
Messages
4
Likes
0
#86
Recognizing that all power dissipation in the core is thermal, I think earlier discussions in this thread of scaling permissible power input from the dimensions of similar chucks with known ratings (and preferably, identical vintage since insulating materials have evolved) is a reasonable way to estimate permissible power dissipation in a chuck of unknown rating. Whether the scaling should be on surface area or volume is beyond me. Starting with a known standard of nearly identical dimensions would be ideal.
 

whitmore

Active Member
Active Member
Joined
Mar 3, 2017
Messages
173
Likes
92
#87
Whitmore, I'm not sure that you will get the desired indication of saturation in your graph if you are relying only on a DC ammeter. ... the desirability of having an AC ammeter in the circuit and the fact that it will indicate saturation, but I think an AC ammeter, not the DC I/V plot, should used as the primary indicator of magnetic saturation.
My concern was that AC ammeters are sometimes RMS, sometimes peak-and-correction, sometimes rectify-and-correction; it's usually unclear what exactly is being
measured (a transformer-coupled oscilloscope would be ideal, though).
Hopefully, all DC meters average over a few tenths of a second.

If the magnet saturates, the current rise time gets faster, so the peak current
happens sooner than when the full magnet inductance was present.
Because inductive current is phase lagged, that means more power transferred
to the load, and should cause a deviation on the I/V curve. I think.

Maybe a true-power meter on the AC input is a cleaner choice?
 

benmychree

John York
H-M Supporter-Premium Member
Joined
Jun 7, 2013
Messages
998
Likes
565
#88
Pardon, if this is not germane to the discussion, but I have a Brown & Sharpe Micromaster Surface grinder that came with a Neutrofier chuck control; when it degaussing feature is in use, one can see the grinding particles on the chuck poles lean one way, and the other way in a diminishing strength as the cycle does its thing; this suggests to me that the cycle is using direct current and diminishing the voltage as each reversal cycle is gone through, until there is no residual magnetism present in the chuck or work piece, at least not a discernible amount. From the noise it makes, it would seem that an electric motor drives a switch that does the current reversal. I also had a Chinese control on a 12 X 36 Thompson grinder; it was solid state and was tiny compared to the Neutrofier . It seemed to work the same way, by reversing DC input in a diminishing series of reversals. Both worked excellently.
 

hman

Active User
H-M Supporter-Premium Member
Joined
Feb 17, 2013
Messages
1,530
Likes
1,045
#89
That sure sounds like what a PTC would do in an AC circuit - reversals of diminishing strength - except that when using AC, the rate of reversal would be too fast to readily discern by watching the particles.

I'm still waiting for the PTCs I ordered (slow boat from China), but as soon as I have them, I'll announce it here and offer them to anybody for just the cost of postage.
 

Bob Korves

H-M Supporter - Premium Member
H-M Supporter-Premium Member
Joined
Jul 2, 2014
Messages
4,032
Likes
4,182
#90
Pardon, if this is not germane to the discussion, but I have a Brown & Sharpe Micromaster Surface grinder that came with a Neutrofier chuck control; when it degaussing feature is in use, one can see the grinding particles on the chuck poles lean one way, and the other way in a diminishing strength as the cycle does its thing; this suggests to me that the cycle is using direct current and diminishing the voltage as each reversal cycle is gone through, until there is no residual magnetism present in the chuck or work piece, at least not a discernible amount. From the noise it makes, it would seem that an electric motor drives a switch that does the current reversal. I also had a Chinese control on a 12 X 36 Thompson grinder; it was solid state and was tiny compared to the Neutrofier . It seemed to work the same way, by reversing DC input in a diminishing series of reversals. Both worked excellently.
Hi, John, welcome to the thread! John (benmychree) is who I got the chuck we are talking about from, and a B & S surface grinder to go along with it. ;)
Edit: John, I will see you Saturday morning at the mill.
 
Last edited:
6
5 7