[harveyjbarrow]

Hi,
I am trying to build a Peltier cooler for my mini-CNC enclosure. I machine tooling board and I do not want the dust to be spread through the room so have built a sealed enclosure. I tried venting and also active venting but the machine still heaps up too much and with the machine being made out of aluminium it is affecting the accuracy of my parts.


I am currently running 4 x TEC1-12706 modules at 70w each. I have tunnel type heatsinks on both sides running both inlet and outlet 80mm fans and controlled with a STC-1000. I have made it 2 stage so only 2 Peltier's come on first and then switches to all 4.

My problem is that it starts fine but after 1 hour it cannot keep up. Also the motors on the CNC machine have no heat output specs. I have the following motors:
Teknomotor Spindle Motor - 330w
4 x stepper motors - 134w each
So that totals 866w but my understanding is only some of that will turn into actual heat.
My next stage was to look at water rather than heatsinks so have a water block with radiator on both the hot and cold side and mount the radiator on the inside.

Any ideas would be most appreciated

[RaderSidetrack]

So that totals 866w but my understanding is only some of that will turn into actual heat.

While "some" of the 866 watts will turn into heat (radiated from the motors & spindle), the balance of the 866 watts will also turn into heat (via friction, mostly between the tool and the work piece.) That is assuming the spindle and motors are being driven to their ratings, actual consumption could be more (or less).

The electrical energy "used" is never destroyed, it is simply converted to another form, in your case, heat.

"The law of conservation of energy states that the total energy of an isolated system remains constant; it is said to be conserved over time." https://en.wikipedia.org/wiki/Conservation_of_energy

[routalot]

Do you have any extraction inside the enclosure?Simply removing air,and hopefully dust,will introduce ambient temperature air into the enclosure and will have some effect.I do know that effective extraction for a 5 axis machine is not easy but if you are running a 3 axis machine,it ought to be possible to collect some of the dust and to push the warm air somewhere else-outside the building is best as it keeps the inside of the workshop cooler.The extraction ought to allow air to reach the tool rather than having it shielded by dust and I have found tooling board to be a very effective insulator.

You don't mention which family of tooling board is normally used and this might be a factor.The polyurethane boards are quite benign but some of the filler powders used in the epoxy board are extremely abrasive and will take the edge off tools in a fraction of the time that they would need in a softer board to reach that level of bluntness.I'm sure you will appreciate that a blunt tool leads to more heat build up and I wonder whether the satisfactory running for the first hour or so is because the tool is still in good shape.

[harveyjbarrow]

Thank you for the reply

The electrical energy "used" is never destroyed, it is simply converted to another form, in your case, heat.

This makes sense, and I will work on solutions that need to remove all 866w which is why I'm failing with my 280w Peltier with its low effiency

[harveyjbarrow]

Thank you for the reply

Do you have any extraction inside the enclosure?Simply removing air,and hopefully dust,will introduce ambient temperature air into the enclosure and will have some effect.I do know that effective extraction for a 5 axis machine is not easy but if you are running a 3 axis machine,it ought to be possible to collect some of the dust and to push the warm air somewhere else-outside the building is best as it keeps the inside of the workshop cooler.The extraction ought to allow air to reach the tool rather than having it shielded by dust and I have found tooling board to be a very effective insulator.

You don't mention which family of tooling board is normally used and this might be a factor.The polyurethane boards are quite benign but some of the filler powders used in the epoxy board are extremely abrasive and will take the edge off tools in a fraction of the time that they would need in a softer board to reach that level of bluntness.I'm sure you will appreciate that a blunt tool leads to more heat build up and I wonder whether the satisfactory running for the first hour or so is because the tool is still in good shape.

I am using polyurethane boards. My machine is a sherline mill and I used to run it in an open environment with the original spindle. The motor blew up so I replaced with a Teknomotor spindle that is air cooled and I want to control the dust in the workshop which is why I am going for a sealed enclosure setup. I did have extraction inside the enclosure using one of my woodworking 2hp vacuum setups but the temps kept on rising as the ambient air got hotter.

My actual problem that I am trying to resolve is accuracy. If I pause a job and re-home at the same time of day no problem, but if I finish a job in the evening and then re-home in the morning I am getting differences of around 0.1 - 0.2mm. My thinking is that the machine (aluminium construction) is expanding during the days running and then shrinking in the cooler conditions overnight. This makes sense in the way that in the morning the machine is cutting deeper so I think that the machine is starting at the correct z axis position but over 8-12 hours is moving away from the tooling board. The parts that I am making are very detailed (0.5mm ball end, 0.1mm step) and take approx. 48 hours of run time.

[joeavaerage]

Hi,
I think the Peltier modules are just overwhelmed. Extracting heat energy is an intensive process.

I bought a new spindle about a year ago, 3.5kW and 40,000rpm. It is water cooled but after protracted running, say 2 hours or more the 20l water coolant would get overwarm and allow the temp of the spindle
to rise higher than I would like.

My neighbour is a refrigeration tech and so he made me a chiller that keeps the coolant water below 20C. The components came out of a largish two door commercial display/cooler cabinet, it works a treat. It will
eventually freeze the cooling water, so it extracts heat faster than the spindle can dump it in.

I did some experiments, pretty rough and ready, but found that the 'heat extraction' ability of the setup is about 450W, while the spindle dumps about 300W into the cooling water.

The point here is that extracting even 450W of heat from a fluid, be it cooling water or a CNC enclosure requires quite a substantial plant,I suspect very much more than four 70W Peltier units.

Craig

[harveyjbarrow]

Thank for the reply!

My current thinking is I will go with this option:

Water Chiller - https://www.teyuchiller.com/water-ch...-co2-lasers_p3
Radiator - https://computerlounge.co.nz/product...opper-radiator

This way I get 1200w of potential heat conversion from the radiator and 1430w cooling from the water chiller. This should be more than enough for the 866w of my setup. Also 2 of those steppers spend most of the time in standby mode ( I use a Gecko G540 driver ).

Also the obvious route here would be just to install an electrical enclosure sealed AC unit but locally here in New Zealand they start at approx. NZD$3,500 and a bit out of my price range.

Any thoughts?

[harveyjbarrow]

Hi all,

Thought I would give an update about the water to air heat exchanger which I have begun testing today.

I have built this with the Teyu CW-5200 1430w water chiller and a 1400w radiator. I also have a STC-1000 controlling fans to maintain the temp in the enclosure.

I have the water chiller set to 20 degrees (C) and the enclosure to 24. I have wired the STC-1000 as a 2 stage setup, so one fan will come on when no heating is required and all fans ( 3 of ) come on when cooling is required. At this stage it has never required all fans to operate.

I have built a separate enclosure that is connected to the side of my main enclosure to house the radiator and fans.

Tomorrow I will do a full 12-14 hours of machining so will be the ultimate test.

Thanks for all of the feedback!!

[joeavaerage]

Hi,
my neighbour is a refrigeration tech and he was going to give the chiller he made to me, but I paid him $200NZD, and even that still felt like I was undervaluing the work he did.

Craig