Shrink Fit....How much induction power?

Hi,
I have recently bought a new spindle which has an HSK32 tool interface. I would like to move to shrink fit tooling, but cannot afford a machine to do so.

Electronics is my thing, so I could entirely probably make my own induction heater. It sort of comes down to power, easy to make something up to 1kW lets say,
but a damn sight harder to build something of 20kW.

Does anyone know, or have experience, about the power levels required to make shrink fit tooling practical?

Craig

https://hackaday.com/2020/10/26/simp...hrink-fitting/

There are DIY shrink fit examples out there.

Think also about how you will:
1. Control tool stick out (not have end mill end up at maximal depth in toolholder)
2. extract the tool safely and without damage

Hi,
I have yet to look at the video but it does not say how much induction power is required. Looking at the size of the coil I would guess around 1kW.

The HSK 32 tool holders have a threaded screw that allows you to set the tool depth. The screw can, at need, be removed so that you could 'push' a broken tool out.
There are commercially available tightening fixtures that grip the taper, without damage. Maybe they could be used.

The bottom line is that induction heaters up to 1kW are not too bad, but at higher power levels the capacitors in particular become harder to get and way more expensive when
you do. I have not committed to making my own induction heater, but I do want to gather some ideas so that I realistically appraise what sort of effort is involved and whether making my own
is practical.

https://www.ebay.com/itm/353090713814

Craig

Looks like it's near 1kw

"The induction heater itself is very simple — a 48 volt, 20 amp power supply, an off-the-shelf zero-voltage switching (ZVS) driver, and a heavy copper coil"

Hi pippin,
yes that's what it looks like....have you actually tried it? Real world experience is what I want.

For instance the current from the power supply was X when I put a 1/2 inch bar in it, but only Y when I put a 1/4 inch bar in it.

Next question is what frequency the thing runs at?. High frequencies are more suitable for surface heating whereas low frequencies are for through heating, this
is a basic characteristic of induction heaters....so what sort of frequencies would be correct for a shrink fit tool setter?

Next question is what is the area of the coil verses the area of the material to be heated. The general rule is that the closer the fit of the material to the coil the more efficient
the coupling of the energy to the material. There must be some clearance otherwise the material would and could short out one or more turns of the coil. So the question is what is
considered normal and what sort of coupling coefficient does that imply?

All the HSK32 shrink fit tool holders I've seen that suit my purpose are all north of $100USD each. If I want twenty of them I need to be confident I can use them successfully.

These are on Ebay at the moment.......Haimer.....a RollsRoyce brand, ordinarily in excess or $300USD each. I would be keen on getting them...but naturally only if I can use them,
and at $65USD are very attractive.

Andrews is another US brand, I'm not familiar with, but look the part. They are for 1/8th tools, and I use a lot of 1/8th tools for PCBs. Getting a bunch of these would be very nice, but
at $129USD each again I need to be sure that I can use them properly.

Craig

Hi,
sorry, the supplier is Andrews but the manufacturer is 'Command'.

Craig

Hi,
I have been experimenting with different arrangements.

The coil needs to be about 40mm in diameter, big enough to easily fit over the tool holder nose, but small enough that there be good coupling from the coil to the tool nose.
The coil needs to be about 40mm long, again to suit the tool nose of HSK tool holders. this suggests about ten turns of 2.5mm/3mm wire (or copper tube) and in turn results in an
inductance of 2.5uH.

I have found and purchased a 4uF 1200V polypropylene capacitor made for induction purposes.

The combination should result in a tank that oscillates at 50kHz....which I think is about the sweet spot for the through heating I want.

I have researched various ZVS circuits, and they have the advantage of simplicity, but am more inclined to use a Hall current detector, or a CT, and use ZCS instead.
Also contemplating using a genuine H bridge rather than a half bridge as is common.

I already have an isolated high voltage, high current, high side MOSFET switch circuit that I use for high and low side switches for servo/motor drives. What may not be apparent
from the pic is that the circuit board used for this switch has a very thick copper layer, 420um or 12oz copper. Thus the high current traces on the underside of the board can handle 48A continuous.
The MOSFET is 650V....so this board was always designed and destined to be used at 400VDC to 450VDC levels and at north of 75kHz.

A genuine H bridge then does not really require any extra work as I already have a design, tested and ready to rock.

Craig