Resurrection of a CFEI 100 KVA Induction Furnace

[awemawson]

I wasn't over happy using 6 mm poly-carbonate to mount the new contactor - it can flex a bit and can shatter - think door being slammed on the handle mechanism - so I decided to remake the mount in 2 mm zintec steel sheet.

Obviously a job for the CNC plasma cutter - so I drew up the panel in Autocad - ported it to sheet cam - and oops - the PC on my CNC plasma table has died

Gives very odd power on self test beeps that don't feature in the book of words. So diversion into PC fixing - it's quite embedded physically but I got it out and apart - no obvious problems - re-seated things, blew it out with an airline and it came back to life. Probably moisture - very damp at the moment in the welding shop.

OK steel panel made - this let me tap it 3/16 UNC to mount the new (to me)  ABB SACE S3 contactor despite having no nuts !

Then a bit of hefty re-wiring in extremely stiff 35 mm CSA 'tails', followed by re-wiring the trip circuit to use the normally closed contacts on the trip delay unit. . . but no . . not so fast . . .the blooming normally closed contacts aren't working - more moisture problems i think. Well it turns out the the base Estop relay that the delay unit fixes to has an unused normally closed pair of contacts - heck who wants a delay anyway !

So putting my fingers in my ears I put 415 volt 3 phase from a local 16 amp circuit on it as I've done previously and yes everything seems to work as it should. Hit the red button and the contactor trips, power up and the controller goes ready, allows me to 'ping' the crucible coil and then I did a VERY brief 'heat' remember only on a 16 amp circuit at the moment.

I think the only thing left to do is sort out a through door operating handle for the contactor - I've failed to source a genuine ABB SACE S3 one but may be able to machine a converter for the original Socomec one - but that's for tomorrow - been at it all day and I've had enough.

[awemawson]

Happily it turns out that adapting the original operating handle from the Socomec contactor to the ABB SACE one proved far easier than I had dared hope

The ABB has a 10 mm square operator bar - the Socomac has a 15 mm square bar, AND the original physical door interlock mechanism (Female slotted arrangement on the door and a nylon unequal rectangle on the contactor) will JUST squeeze into the depth where the new contactor sits allowing not only electrical operation of the trip switch but also mechanical interlocking preventing the door being opened when the switch is 'on')

So by milling part of the 15 mm operating bar down to 10 mm square using a rather flaky clamping set up I was able to re-assemble it all AND it works !

I had envisioned having to create 3D parts and printing them out but hoo-ray not needed and the furnace is now back in commission.

[russ57]

Now just need to remember why you needed it working...

[awemawson]

Russ - next job is to cast a bit of bronze into a bar for Pete. to turn into an anti-backlash nut for his mill, however external pipework is frozen at the moment !

I had a bit of luck on eBay where an identical 'delay timer module' to the one that I had to remove from the eStop circuit turned up at a very modest price. It's been rather held up in the post, but  arrived today and it was rather TOO identical in that it's normally closed contacts were open circuit at all times  testing with a Fluke

However it had obviously been previously only using the 'normally open' contacts in it's former application so the other pair had probably been oxidising gently for the last thirty years. Increasing the wetting voltage to 24 on a lab supply with current limiting almost immediately it started working OK and now is fine. (In my application if this contact is not closed the machine won't connect to mains so any doubt about the contact isn't a safety issue)

These units are not designed to be taken apart otherwise a simple clean with a bit of nice abrasive printer paper between the contacts would have given the same result.

When things are a bit warmer in the foundry I'll fit it!

[awemawson]

A Fast Foray -into a Freezing Foundry and I've Fitted the replacement time delay.

First a quick picture for the record of how I'd borrowed a spare normally closed contact pair on the host eStop contactor in case things didn't work out, then a simple job of re-routing the wiring in the panel conduit having clipped on the time delay. Testing it works just as expected so thankfully a 'none event'.

I was a bit puzzled why the designer had incorporated this 300 mSec delay but thinking about it it gives the microprocessor 300 mSec to close down the various drivers etc before power is lost so probably sensible.


. . .so no 'guv we was never there . . it's exactly as it was 

[awemawson]

I decided to re-fit the safety covers that I moulded in fibreglass many years ago to cover up the exposed electrical connections to the pot part of the furnace - I couldn't remember why I'd not put them back - now I realise !

It turns out that when I decided to put heavy cable strap links across the quick disconnect fittings as their snap balls were suffering from the huge currents that they were carrying I fitted the straps in such a way that they foul the cover  It can be re-worked I think fairly easily but not today - my fingers are falling off having done the little I've already done !

[awemawson]

It turned out that the 'modification' was far less of a problem than I'd feared - merely having to shorten the 70 mm CSA cables by about two inches and having ensured that both lugs on a link were mutually at 90 degrees fix the outer one to the underside of the wide copper strip rather than the top. This way the 12 mm bolts no longer fouled the case nor did the link loops.

Altogether much more satisfactory and certainly safer than it has been. There's enough to keep an eye on when pouring hot metal without also risking inadvertent electrocution !

So it's now as I'd intended to years back. I had the pattern I'd made for the fibreglass cover kicking about for a long time but I suspect that it's been disposed of now.

[hermetic]

Looking good Andrew, its coming together well! I am  a bit ticklish on high volts/current today. I used to think nothing about terminating up to 600V at many hundreds of amps, but it is something I have found as I get older that I double and treble check insulation barriers and seperation.Even installing my new 3 phase service panel made me a little nervous after I have been out of the industry for so many years, and I have a huge 3 phase steam cleaner which I "haven't got round to testing" after rebuilding it!!
Phil

[awemawson]

Thanks Phil for the kind words. I made the cover when I was still in Bromley so that's at least 15 years ago. Trouble is not only is this high (ish - 600 v) voltage but it's about 3 kHz with spikey fast rising edges so it doesn't behave like normal 50 c/s stuff !

Cast your minds back to when the feeder hose / cables were resting too close to the stainless steel braded air hoses, and the coupling induced so much current to set them on fire 1

Post #756 here : https://www.madmodder.net/index.php/topic,13100.msg160605/topicseen.html#msg160605

So my next task is to make something a bit more suitable than the offcut of floor board that I'm currently (sorry for the pun) using !

[hermetic]

it's about 3 kHz with spikey fast rising edges so it doesn't behave like normal 50 c/s stuff !
 

Narsty!!
Phil

[awemawson]

Not seen it yet, but a friend (at my prompting) reckons he's found an odd length of cement reinforced downpipe approximatly 4" O/D, and my calculations indicate that if I mount this horizontally on the 'casting body' frame the Brewers Hose cables can rest on it keeping them off the air lines.

All theory at the moment - I did find a bit of 5" i/d 5.5" o/d flue pipe  but it's a bit too large and kinks the pipes.

[awemawson]

Today was the first opportunity to do a 'full power' test on the new breaker arrangement. Cottage guests were out so I could run the generator, and it was no longer freezing so I could return the bore hole pump to service.

So turning on the bore hole pump I opened the ball valve allowing water the flow through the heat exchanger to be presented with a blast of high pressure water all over the place - more frost damage on the far side of the ball valve so at least 'turn offable' !

OK it needs repairing but press on - the water chiller has a few gallons in the tank of glycol dosed water that it circulates through the furnace, so it can run for a short while without the heat exchanger.

I ran the furnace up to about 60 kW and all went well just heating a bit of 20 mm re-bar to glowing but then stopped before things over heat without the bore hole water,

Digging through pictures that I took installing the plumbing that's sprung a leak it's not the easiest to get at but possible - so that's my next task in the next few days - and perhaps extend my frost trace heating system.

[awemawson]

Frost prevention measures:

My first thought had been the wrap snti-frost tape round the two umbilical hoses that join the water chiller to the flow and return mainifold to the bore hole.

Now tape isn't particulary convenient on flexible pipe but it can be done.

Then I thought: the return pipe is open at the well end - can I adjust the piping so that opening a ball valve it all flows out. But no, the height of the manifold compared to the heat exchanger prevents this.

My current thought is, if when I make good the frost damage on the flow pipe I incorporate  an air input valve, when cold temperatures threaten I can maybe blow the water from both the flow, return and heat exchanger back down the return pipe (thus overcoming gravity) leaving hopefully minimal water in the system.

Any other suggestions would be very welcome.

[awemawson]

Today I stripped down the water feed  and return pipes from the Borehole to let me take the panel off the wall in the foundry and investigate what actually has burst.

To my surprise I've found that the actual body of the ball valve has split - it's a dead ball valve 

So having ordered up a replacement I've also decided on a 'frost strategy' for the future. The pipe rises from the ball valve and goes through 90 degrees to a convenient  place on the panel. I will replace the 90 degree 1" BSP elbow with a Tee - and the extra 'way' will be reduced to 1/4" BSP, have a 1/4" Ball Valve followed by a 'Male PCL' air fitting.

This scheme should allow me to 'blow down' water from the pipe work and the side of the heat exchanger that isn't antifreeze filled.

As an extra line of defence I've also ordered decent quality pipe lagging and am off now to hunt for a suitable length of frost stop heat trace cable.

[awemawson]

Glad to say UPS delivered my B.E.S. plumbing order nice and early so I was able to get on with putting it all back together.

A bit of heat released the Loctite holding the 1" BSP elbow to the various other fittings and having replaced the elbow with a Tee, the ballcock with one without a burst bit, and installed the 1/4" BSP air ball cock and PCL fitting I was able to put it all back on the wall. I took great care to mark the blue MDPE pipes for insertion depth and seat them firmly as both have to go on as an assembly - if I've got it right it wont leak . . time only will tell !

Then I removed the old nylon reinforced umbilical pipes that take the feed from the borehole to the heat exchanger, and the return that goes eventually to the stream via our land drains. I had bought some much more substantial stainless steel wire reinforced rubber 'Neptune' hose quite a long time ago as the nylon was a temporary solution - it had worked OK but wouldn't take kindly to a splash of molten metal. So now was the time to uses it.

All went well and I'm now waiting a few hours for the Loctite 542 to cure before I pressurise and test it.

The split / self fixing 1/2" wall insulation for these pipes has also arrived but the trace heating cable hasn't so I can't wrap up the umbilicals just yet. It's going to be great fun installing the trace heating. I've ordered a single  14 metre length the mains 'start of which will begin at the wall panel, wrap in a spiral round the return pipe, enter the chassis of the chiller wrapping round its internal pipes and exit wrapping around the flow pipe ending back at the panel. Or that's the theory - I can imagine quite a tangle during the wrapping !

[awemawson]

So I put on my brave trousers this morning and wetted up the system - great relief - no leaks phew !
Amusingly I noticed that the water from the bore hole was warmer initially than that circulating in the (not powered up) furnace body and electronics driver (10C compared to about 6C)

So today I've been routing the trace anti frost wire that turned up this morning - it was every bit of the nightmare that I'd expected. No where could I find a minimum bend radius for it - wrapping round the heat exchanger I'd have liked to pull it much tighter but didn't dare.

Lagging the heat exchanger was the obvious thing to do, and I've masses of rock wool slab left over from quietening the generator, but fixing it was a pain. Initially I thought dabs of hot glue gun glue, but fibres just pulled off the slab. Then I thought ty-wraps - which sort of work but if pulled to a sensible tension just squash it, so it's not as neat as I'd wish.

Tomorrow I'll embark on fitting the split lagging to the pipework (Note the pipes with no trace heating are the furnace side of the heat exchanger so are heavily dosed with glycol antifreeze)

[awemawson]

This morning I got the pipe lagging fitted.

The pre-applied sticky stuff on the split lagging is pretty effective but is a menace as you try to guide it into tight spots.

I decided to back it up with ty-wraps as I don't quite trust it not to move over time, especially on bends.

[awemawson]

Before pushing the chiller unit back against the wall I thought I'd fit a foil jacket around  the heat exchanger for two reasons. Firstly the lagging is far from continuous at the top so the marginally warmed air from the trace heating / anti frost tape will rise and go to waste, and secondly the insulation being rock wool is rather friable and tends to shed fibres.

I used standard 'kitchen foil' which is terrible to wrap round things like this with sharp bits - it tears at the slightest catching. I'm sure normal aluminium foil used to be much thicker - this measures as 10 microns - there must be better stuff available surely.

Anyway hopefully that's the frost damage repairs and anti-frost measures finished at long last.

Having dragged the chassis out and pushed it back a few times I thought it sensible to run it for a while to prove it was all still water tight - yes I'm glad to say. Water was coming from the bore hole at 11 degrees C and it took quite a while for the equipment to rise from the 7 - 8 degree level up to marginally below 11. Potential ground source heat pump I suppose !

Anyway maybe I can now get back to casting . . .

[ddmckee54]

Andrew:

You've got a nice little box shaped bundle there.  Maybe you should check with an HVAC installer and see if you could get some of the foil backed insulation board that they use.  You could use it to make yourself a nice little removable insulated box to cover the heat exchanger.

Don

[awemawson]

Nice idea Don but the rockwool and kitchen foil were 'in stock' ! The four pipe and the bolting to chassis also wouldn't help, but it's now done and I'll see how it performs when we have the next cold snap.

The heat cable thermostat turns on at about 3 - 4 degrees C and with a bit of hysteresis probably goes off at 6 or so.

[tom osselton]

 

[ddmckee54]

Andrew:

I totally agree, with all the protrusions it would be a royal PITA to fabricate.  Something like that might be something to keep in mind though, if after the next cold snap you decide the heat exchanger needs a little less coolth and a little more warmth.

Don

[awemawson]

I actually USED it today - first time in months !

It was just a proof of concept sort of thing as I sort out the logistics of pouring from the tilting body single handed. I'd like to be able to pour into a conventional pouring ladle and thence into the mould but that implies pre-heating the ladle which has to be supported in the right place catch the charge from the tllting crucible. At the moment the whole furnace body is raised 9 inches on a dolly for convenience of moving about but that makes looking into the 'pot'  a bit hard.

Plan is to put it in it's final operating position off the dolly and weld up a frame to hold moulds,   ladles etc to sit on the floor infront of it. As it is at the moment the melt is up at chest level and leaves me very much in the danger zone when pouring. Once the melt is safely in a ladle it can be distributed to more than one mould on the floor.

But before then I need to clear a path from the Tractor Shed, via the Stable to the roller shutter door on the foundry so I can lift the furnace body off it's dolly with my forklift truck. There is a Zed purlin above it, and I got a block and tackle out to lift it but bottled out thinking bending the purlin and bringing the roof down probably isn't a good idea!

Still I did turn half a huge lorry piston into an ingot today

Oldish Video of the furnace body tilting to see what I'm on about

[Pete W.]

Andrew,

Congratulations on the progress you are making.

Have you studied how the centre of gravity moves as you tilt the machine with a full molten load?  The legs look to me as though they don't extend forward far enough.  When you get it finally positioned and off the dolly, shall you rawlbolt the legs to the floor?

[awemawson]

Thanks Pete for the kind words.

Remember that it's sitting on a slightly wobbly dolly at the moment. I probably won't bolt it down but will be adding significant weight in the form of some flag stones sitting on the base. I may change my mind though when I've used it with a heavier charge