1. Home
  2. Electronics Design
  3. Ringing and coolant.

Ringing and coolant.

Why is it when you try so hard to reduce or eliminate induction ringing, when it so prevalent that one day you decide that you actually need it and it won't ring!

I do some moon lighting on the side and yesterday I got a call from a person that has a small machine shop in his garage. He has some tool&Die machines that in the final process of the dies for example, the CNC operation turns the dies or puts the finishing cuts on it then what happens the collet releases or what ever it uses to hold the piece, then gets transported just forward of that where there is an induction coil.

This is where I come in, I have updated this machine with new control electronics because it was old and it has IR sensing, quenching oil spray etc,that all has to be monitor and regulated. I integrated a new micro controller in it with some updated electronics.

This is the problem, he had to get some more coolant for the induction part. This coolant goes through the tube etc.. The induction inverter is actually a ringing high Q with a low Z pulse operating the circuit with low ESR caps etc.. He tells me the new supply of coolant is causing the coil to ring at a much reduce rate.. I have a monitor circuit in there to make sure it does ring incase the fittings are loose, caps get bad or some termination is going south..

The induction coil is of 1/8 copper tubing and it should be ringing around 50khz. He's telling me the monitor is report a very low ring and he can see this in the heating for the work piece as the IR sensor is reporting a much longer heating time. The coolant passes through the coil.

Is it possible he got some coolant not designed for this operation ? All I can think of is maybe he is using something with mercury in it?

He cleaned out the tank and put in some older fluid he had and it now works as it should.

Some suggestions/ideas of this effect would be greatly appreciated, I am not a chemist... :(

And SLow-Man, STFU!

Jamie

Reply to
Jamie
Loading thread data ...

rol

n
g
.
?

now

Jamie is as confused as ever. If he had two neurones to rub together, he might at least have told us what the coolant was supposed to be.

My guess is that the induction coil is a loop of copper tubing, with an 1/8" outside diameter, and the coolant is pumped through it. The coolant might be water, probably dosed with chemicals to stop bugs growing in it and to prevent or slow down rust in the steel bits of coolant circuit. Water might be a bit viscous for a 1/8" outside diameter copper tube, but anything else would be a lot more expensive and a lot more trouble to look after.

What he means by "ringing" is equally obscure. Presumably the copper coil is the inductor in tank circuit with the low ESR caps, and it gets banged with a narrow spike of current from time to time, which excites a exponentially decaying sinusoidal oscillation in the tank circuit, damped by the the resistive losses in the copper coil and the work-piece being warmed/annealed.

The only obvious way for the coolant to kill the ringing - as in speed up the exponential decay - would be for it to fail to cool the copper coil, which would get hot and consequently have a higher resistance, but it would have to get very hot to make the kind of difference that somebody like Jamie would notice.

If the coolant gets hot enough to boil inside the copper coil, heat transfer initially increases like crazy - sub-cooled nucleate boiling

- but drops way below its initial value as the surface gets even hotter.

formatting link

Maybe the induction coil cooling was designed on the basis of sub- cooled nucleate boiling and the new coolant messed that up.

-- Bill Sloman, Nijmegen

Reply to
Bill Sloman

Jamie schrieb:

Hello,

I cant imagine an effect of the coolant inside the copper tube on the inductance of the coil.

May be he changed something else when changing the coolant.

The heated material inside the coil is the same as before, with the same dimensions?

Bye

Reply to
Uwe Hercksen

I can't imagine that you could buy coolant with significant amounts of mercury in it without having all sorts of issues with the EPA.

Water-based coolant? At any rate, if the coolant is conductive it'll be a Q-killer -- water based or not. The coolant must be flowing through metal stuff that's grounded, and it's flowing through the electrically hot side of the coil. If the coolant is conductive, it'll just act like a resistor from that coil 'hot' side to ground.

I'd either check the coolant specifications for conductivity, or measure conductivity directly. You can make a rough check just dipping meter leads into the mix, but you can make better measurements if you make yourself some sort of jig involving parallel plates, to get yourself more area with more control over spacing.

--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com
Reply to
Tim Wescott

,

me

e
n

il.

ntrol

ion

s
r

ing

on ?

t now

I
e

Not if it is pumped around inside a copper tube - the conductivity of the copper is going to be way higher than the conductivity of any coolant (short of mercury or Wood's metal).

-- Bill Sloman, Nijmegen

Reply to
Bill Sloman

ng,

t

some

the

at

hen

coil.

control

w

ction

is

th

he

e

or

nging

nd

tion ?

it now

d, I

be

It's entirely possible that the coil is on the hot side of the circuit, with some dielectric break in the coolant line. (piece of rubber hose) So that a conductive coolant is causing more R to ground... killing the Q. (The coolant pump is most likely grounded.)

George H.

Reply to
George Herold

ging,

it

ll

s some

, the

what

then

n coil.

w control

new

duction

er is

with

the

ere

ad or

ringing

and

is

e

ration ?

t?

nd it now

ted, I

f

ll be

Only physicists build things like that - and I've seen them do it. Regular mortals have more sense, and less access to conductivity water and the instruments to measure its conductivity.

-- Bill Sloman, Nijmegen

Reply to
Bill Sloman

Let's say the coolant was oil. The copper coil has tails which connect through (insulating) Silicone tubing to a radiator, a recirculating pump, and a header tank. I've seen similar for heating fatigue test specimens.

Replace the oil with water and you put a resistance across the coil.

Cheers

--
Syd
Reply to
Syd Rumpo

What happens is, the amplitude of the high Q ring is much lower when he is using this coolant he just got, it takes longer for the work piece to heat up. THere is a IR sensor pointed at the work piece when it pops out of the holder. when it reaches a set point, it then has a flat time curve it must sit there before it gets past on to the next step where quenching is then taken place and that is also timed to get the hardness just right.. at the end it drops in a well to be collected after it gets filled.

He also told me that this same fluid does basically the same thing on another machine that still has the original electronics in it. So it is something that fluid is doing which is making that coil not behave properly.

I had him do a temp check on the termination's and coil fittings, they all seem to be ok.

I love mysteries, some times the answers lead to new idea's for use :)

Jamie

Reply to
Jamie

I told you not pto reply! The coolant is in a chiller tank with refrigeration. it does not get hot! It needs to stay cold so the coil can operate at it's highest Q. The temperature of the fluid is staying down where it belongs, which is cold! I've already checked this.

Jamie

Reply to
Jamie

That is my take on it, I need to get a sample of this and do a check. I had him to a R check which the tank has that in it in case it gets dirty, which is why he changed it in the first place. The indicator was reporting a conduction in there however, the circuit is DC for this and I am wondering if maybe the 50Khz is some how interacting in there, like maybe it has some magnetic properties that only appear under AC ?

He got a 50 gal drum of this stuff with no option to return it. I think he'll end up thinning it and using it as is until depleted.

Jamie

Reply to
Jamie

I wish that to be true how ever, I would then have a problem connecting to it electrically :) It does have electrically isolated coupling for the fluid lines.

Unless of course, the fluid is very conductive and the isolation isn't far spread enough? There is a 1 inch isolation coupling there. Maybe it isn't long enough. I guess I could have him try using rubber hoses all the way to the pump and return side... Hmm...

You may have something there, good thinking! :)

Jamie

Reply to
Jamie

Well, then, it must have been designed by physicists!

You never gave us estimated impedances of coil and cap, nor an estimate of the conductivity of the coolant, or, for that matter, the inner radius of that 1 inch isolation coupling.

--
My liberal friends think I'm a conservative kook.
My conservative friends think I'm a liberal kook.
Why am I not happy that they have found common ground?

Tim Wescott, Communications, Control, Circuits & Software
http://www.wescottdesign.com
Reply to
Tim Wescott

it depends on the size of the item he is machining, the coils can be selected. He has a shadow board of them.

But, going from memory, I think the largest one is 3 inches cross section with only about 3 turns on it. It seems the smaller coils have more turns. Most likely to keep induction with in a range.

The coil is selected verses the size of the work piece and normally there is only a small gap in there like maybe no more than a 1/4 between the work piece and coil.

Induction changes due to the work piece of course, which is why the coil needs to be sized. the low ESR, high ripple caps do have a bank selector..

But if you read my latest post, I think I may have found something but now, how to work with it? This guy can not return this 50 gals of coolant.

The circuit is a combine of the work piece in the coil and caps which will yeild a frequency of 50Khz up to 75khz. It varies, especially when the steel starts to get red.

If more data is needed I can most likely get the whole circuit on a post. I just need to get my hands on the machine manuals which are sitting in the machine's cabinet.

All I had to do in that area was to replace some large caps, power supply work and one blown up hockey puck SCR.

The rest of it was just process controls with a touch screen micro controller system I adapted with all the other things on the machine.

Jamie

Reply to
Jamie

ng,

t

some

the

at

hen

coil.

ntrol

w

ction

is

th

he

e

or

ing

nd

on ?

t now

d, I

l be

sn't

Ahh, it was Tim's idea. I was just suggesting how a conductive fluid might spoil the Q. Is it a water based coolant?

George H.

Reply to
George Herold

rol

ion

s

or

g

il.

?

now

I

You did. Do you think I care whether you want my input or not?

It won't work if it doesn't get hotter when it goes through the coil. You'd need to check the temperature of the coolant at the point where it comes out of the coil - not after it's been mixed back into the contents of the chiller tank.

-- Bill Sloman, Nijmegen

Reply to
Bill Sloman

You want to think about what you just said?

Jamie

Reply to
Jamie

e

trol

w

on

is

th

he

e

d or

ng

nd

coil.

n ?

now

I

I know that you don't want to think about what I've just said, but it's elementary thermodynamics.

If the induction coil was part of a heat pipe, this wouldn't be true. The heat would be being absorbed by the vapourisation of the coolant on the inner surface of the induction coil, and would be dumped in the coolant tank when the vapour re-condensed, but the system you are talking about, with it's range of interchangable induction coils, couldn't work as a heat pipe - you've got to keep the system free of non-condensable gases, otherwise the vapour doesn't get back to the coolant tank fast enough to shift an appreciable amount of heat.

-- Bill Sloman, Nijmegen

Reply to
Bill Sloman

Your lack of familiarity with resonant circuits is showing - again.

AFAICS from the description, the coolant, (took me a while to get on Jamie's wavelength, since he was talking about a CNC machine, for which "coolant" means something different to me), is pumped along the interior of a coil made out of copper tube, fed in via insulating pipes or hoses. If the coolant is conductive, this will produce an effective resistance in parallel with the coil, lowering its Q. If the coil is part of a resonant circuit,the Q of the whole circuit will be lowered, and, for a given excitation, the resonant amplitude will decrease.

That'll speed up exponential decay all right!

Mike Terrell will doubtless have come across similar problems with contaminated cooling water in big water-cooled transmitters.

My solution (no pun) to the problem would be to use plain deionized water, use one of those handheld combined pH and conductivity meters, and change the coolant regularly.

--
"For a successful technology, reality must take precedence 
over public relations, for nature cannot be fooled."
                                       (Richard Feynman)
Reply to
Fred Abse

Fred, I know how the system works. I am sorry that you believe I don't have and understand of basic 101 res circuits, I do. I rebuilt that machine electrically from a pile of parts he had.. Updated the controls etc.. I also wound a couple of coils for him and had them send out for inner plating..

The problem has been solved. The coolant was the wrong type for the machine he uses. This coolant they sent him has an additive for systems that insert plates into the tank for density measurement.

He can't return it because it was his fault for not specifying it however, It has been fixed. I gave him a plastic heat exchanger and he plumbed it in. It is working fine now..

Thanks for your exceptional wisdom, if that is what you call it.. ;)

Jamie

Reply to
Jamie

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.