Anybody here?

[Peter Fairbrother]

A belated Happy Christmas and a better 2021.

Anybody got any ideas for high temperature seals? I want to build a
small vacuum annealing furnace, up to about 1150C.

The inside will be 4x4 steel box section with one welded end. My problem
is the other end, which has to open but be sealable to vacuum.

I don't want to use an outer cool sealable box (outside the insulation).

Any ideas?

thanks,

Peter Fairbrother

[Richard Smith]

Could the easiest by - weld a lid / cap on it?

Grind end of tube to silver, put "lid" there - protruding "plug" you
corner-joint fillet? - and weld MIG/GMAW at a lowish Amps (say 100A
depending on tube thickness - so fuses to clean metal but very little
penetration - so easy to sever when done?

Might be easiest...
Not ideal solution if you are doing this 2 shifts 6 days a week 10
minute cycle time :-) but for a few-off, might be the readily done
reliable one?

[David Billington]

This page suggest a nickel seal might be up to the job 2200F is 1204C.

http://blog.parker.com/use-a-metal-seal-for-those-high-temperature-or-high-pressure-situations

[John]

Grind open end and cover absolutely flat, then use annealed copper
gasket between the two ?. Aka, cylinder headd gasket material.

Depends on how much vacuum leakage and absolute pressure is needed.
You may be able to tolerate a small leakage if the right vac pump is
used...

John

[John]

Of course, rethink this and use an inert gas rather then a vacuum ?...

[Peter Fairbrother]

On 11/02/2021 16:09, John wrote:
> On 02/11/21 16:04, John wrote:
>> On 02/10/21 16:14, Peter Fairbrother wrote:
>>> A belated Happy Christmas and a better 2021.
>>>
>>> Anybody got any ideas for high temperature seals? I want to build a
>>> small vacuum annealing furnace, up to about 1150C.
>>>
>>> The inside will be 4x4 steel box section with one welded end. My problem
>>> is the other end, which has to open but be sealable to vacuum.
>>>
>>> I don't want to use an outer cool sealable box (outside the insulation).
>>>
>>> Any ideas?
>>>
>>> thanks,
>>>
>>> Peter Fairbrother
>>
>> Grind open end and cover absolutely flat, then use annealed copper

>> gasket between the two ?. Aka, cylinder head gasket material.

Not a bad idea.

>>
>> Depends on how much vacuum leakage and absolute pressure is needed.
>> You may be able to tolerate a small leakage if the right vac pump is
>> used...

Would prefer not to have to use continuous pumping.

>> John

> Of course, rethink this and use an inert gas rather then a vacuum ?...

A disposable [1] argon cylinder is about £15, a regulator to fit can be
had for about £10 (and I have >1 of both in my box of stuff).

Still needs some sort of seal, as the regulator gives a few psi, but
less stringent.


BTW if you are wondering why I don't want to use an over-insulation
gas/vacuum containment, it's because a) it's expensive - a pressure
cooker is the best solution I have found, and large tall pressure
cookers are rare and expensive - and b) I want to use Kanthal A1 for the
heating element, which needs an air/oxygen environment to maintain the
Al_2O_3 coating.

Kanthal wire would be wrapped over an electrically insulating thermally
conducing layer over the steel cross-section. Maybe glass tissue /
ciment fondu or ceramic paper / filled 1700C HT cement for the
insulating layer.

Type N thermocouple (like type K but better long-term stability) with
PID controller - hmm, anyone got any ideas for a cheap programmable
ramping PID?


Peter F

[1] they say DO NOT REFILL in big letters 0 but they have a handy
refilling valve..

[John]

On 02/11/21 19:21, Peter Fairbrother wrote:
> On 11/02/2021 16:09, John wrote:
>> On 02/11/21 16:04, John wrote:
>>> On 02/10/21 16:14, Peter Fairbrother wrote:
>>>> A belated Happy Christmas and a better 2021.
>>>>
>>>> Anybody got any ideas for high temperature seals? I want to build a
>>>> small vacuum annealing furnace, up to about 1150C.
>>>>
>>>> The inside will be 4x4 steel box section with one welded end. My
>>>> problem
>>>> is the other end, which has to open but be sealable to vacuum.
>>>>
>>>> I don't want to use an outer cool sealable box (outside the
>>>> insulation).
>>>>
>>>> Any ideas?
>>>>
>>>> thanks,
>>>>
>>>> Peter Fairbrother
>>>
>>> Grind open end and cover absolutely flat, then use annealed copper
>>> gasket between the two ?. Aka, cylinder head gasket material.
>
> Not a bad idea.
>

If you have 4/4" square cover, that's (correct me if i'm wrong), about
160 odd libs/sq" pressure across the face of the cover, which should
seal pretty well.

Question is: do you really need high vacuum or not and if so, why ?.
There will be some leakage, so continuous pumping will almost certainly
be needed.

Need to define requirements a bit more tightly and fill in detail...

John

[David Billington]

I was going to suggest an Arduino as there are a few PID temperature
controller projects out  there and Adafruit produce a thermocouple
module that supports a range of thermocouple types and has a support
library but you can buy cheap Chinese PID controllers off ebay for
around £35 and even TCDirect have a budget range sub £100 IIRC so I
wouldn't bother making my own.

Copper as a gasket isn't going to cut it as it melts below your target
1150C temperature.

[et...@whidbey.com]

If you can use a circular opening then why not machine a taper seal? I
wouldn't be surprised if one of the high temp anit-seize compounds
would be good enough to lube the tapered surfaces.
Eric

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

[Peter Fairbrother]

On 11/02/2021 23:21, David Billington wrote:
[...]

> I was going to suggest an Arduino as there are a few PID temperature
> controller projects out  there and Adafruit produce a thermocouple
> module that supports a range of thermocouple types and has a support
> library but you can buy cheap Chinese PID controllers off ebay for
> around £35

more like £80 - the £35 ones are not ramp programable, even though they
claim to be.

and even TCDirect have a budget range sub £100

£153 4 setpoints, no ramp rate control.

IIRC so I
> wouldn't bother making my own.

Adafruit MCP9600 TC module is £20 or so.. :(

Peter

[David Billington]

On 12/02/2021 01:26, Peter Fairbrother wrote:
> On 11/02/2021 23:21, David Billington wrote:
> [...]
>> I was going to suggest an Arduino as there are a few PID temperature
>> controller projects out  there and Adafruit produce a thermocouple
>> module that supports a range of thermocouple types and has a support
>> library but you can buy cheap Chinese PID controllers off ebay for
>> around £35
>
> more like £80 - the £35 ones are not ramp programable, even though
> they claim to be.

I wonder if it's like a friends PID controller which is a cheap Chinese?
Brainchild? IIRC or similar, like
https://www.peaksensors.co.uk/component-store/controllers/btc-9090-temperature-controller/
, where the controllers programming doesn't actually match what the
documentation for the model says, looks somewhat similar to the cheap
ones on ebay. Annoying in that the ramp rate is in degrees per minute
rather than per hour which is more common in my experience.

[John]

Sounds like overkill to me, as if there's enough thermal mass in
the system, the temp will ramp up and down slowly anyway, even
using a standard on/off style controller. Don't make it more
complex than it needs to be, even if writing the code on a micro
is fun :-).

I usually look out for Omron 40mm EC5xx type controllers for such
applications. Not too expensive s/hand on Ebay and some have direct
dc control outputs to drive solid state relays, which avoids a mech
relay...

John

[Peter Fairbrother]

On 12/02/2021 02:26, David Billington wrote:
> On 12/02/2021 01:26, Peter Fairbrother wrote:
>> On 11/02/2021 23:21, David Billington wrote:
>> [...]
>>> I was going to suggest an Arduino as there are a few PID temperature
>>> controller projects out  there and Adafruit produce a thermocouple
>>> module that supports a range of thermocouple types and has a support
>>> library but you can buy cheap Chinese PID controllers off ebay for
>>> around £35
>>
>> more like £80 - the £35 ones are not ramp programable, even though
>> they claim to be.
>
> I wonder if it's like a friends PID controller which is a cheap Chinese?
> Brainchild? IIRC or similar, like
> https://www.peaksensors.co.uk/component-store/controllers/btc-9090-temperature-controller/
> , where the controllers programming doesn't actually match what the
> documentation for the model says, looks somewhat similar to the cheap
> ones on ebay. Annoying in that the ramp rate is in degrees per minute
> rather than per hour which is more common in my experience.

No, they are just plain ordinary PID controllers with no process step
functions at all.

I think the rationale goes like this: seller pays <£3 (Shenzhen market
price) for a cheap plain PID like the REX100, sells it as a programmable
PID for £35.

Buyer probably notices it isn't a process control PID, but if not
whoopee. If buyer notices, seller offers a 20% discount - if refused, in
order to avoid the bother of shipping it back, seller will offer a 50%
discount.

Maybe even offer a 70% discount - seller still makes a profit!!


Peter

[Peter Fairbrother]

On 12/02/2021 11:28, John wrote:
> On 02/12/21 01:26, Peter Fairbrother wrote:
>> On 11/02/2021 23:21, David Billington wrote:
>> [...]
>>> I was going to suggest an Arduino as there are a few PID temperature
>>> controller projects out  there and Adafruit produce a thermocouple
>>> module that supports a range of thermocouple types and has a support
>>> library but you can buy cheap Chinese PID controllers off ebay for
>>> around £35
>>
>> more like £80 - the £35 ones are not ramp programable, even though they
>> claim to be.
>>
>> and even TCDirect have a budget range sub £100
>>
>> £153 4 setpoints, no ramp rate control.
>>
>>
>> IIRC so I
>>> wouldn't bother making my own.
>>
>> Adafruit MCP9600 TC module is £20 or so.. :(
>>
>> Peter
>>
>>
>
> Sounds like overkill to me, as if there's enough thermal mass in
> the system, the temp will ramp up and down slowly anyway, even
> using  a standard on/off style controller.

I want process control - setpoints, soaks, ramps.

For instance, for investment castings I want to dewax for an hour or two
at 90C, dry at a ramp from 100-250C over several hours, then ramp up to
casting temperature of maybe 700C fairly fast, sound an alarm and hold
'til the melt is ready.

If I have to sit and watch over that I am not going to do it properly. A
process control PID would do a much better job. And it would waste my
precious time.

Don't make it more
> complex than it needs to be, even if writing the code on a micro
> is fun :-).

I hate writing code. That's why I am a cryptologic theoretician not a
cryptographic programmer.

> I usually look out for Omron 40mm EC5xx type controllers for such
> applications.

Why? They are just plain ordinary controllers, no process control
functions.

£300 new, £70 used. Ouch.

A REX-100 for £10 new will do the same job, just make sure to choose the
right model.

Which is why I don't like paying £20 plus for a TC module which only
does a small part of the job.


Peter Fairbrother

[John]

Lol, if you had said to start with you needed a complex control function
for this, then you would have got different answer. Virtusl project
requirements drift, etc :-)...

John

[Peter Fairbrother]

> for this, then you would have got different answer. Virtual project
> requirements drift, etc :-)...

The main - only, really - requirement is to be able to change the SV
over time. A small extra requirement, for something which has the
complexity of PID functionality already built-in.


Peter Fairbrother

[RustyHinge]

On 10/02/2021 16:14, Peter Fairbrother wrote:

> A belated Happy Christmas and a better 2021.

Reciprocated.

>
> Anybody got any ideas for high temperature seals? I want to build a
> small vacuum annealing furnace, up to about 1150C.
>
> The inside will be 4x4 steel box section with one welded end. My problem
> is the other end, which has to open but be sealable to vacuum.
>
> I don't want to use an outer cool sealable box (outside the insulation).
>
> Any ideas?

Sorry - blind Bambi.

--
Rusty Hinge
To err is human. To really foul things up requires a computer and the BOFH.