Selinoid heat

[pncdan]

I have a selinoid coil that i am trying to understand. The coil is copper wire wrapped onto a plastic bobbin and overmolded. A steel shell is place around the coil. The coil assembly is place on a steel valve body and then placed in an oven at 85 Deg C. The selinoid is powered with 16 vdc and allowed to stabilize. I calculate a core temperature of 204 DEC C. The valve body is mounted to a aluminum block. The coil is about 1.250 long and 1.300 dia. I understand some of how the selinoid desipates heat to survive, but what manner does it use the most?. Is heat transfer to the steel core valve and then to the AL the best path because of their conductivity or is more heat shed off of the plastic overmold surface. Air movement is minimal. If the air is a poor conductor of heat verses the steel envolved does this mean the valve is the path of choice for heat transfer ?

dan

 

[sailoday28]

HOw did you determine the core temperature?

 

[ko99]

Sailoday's question is a good one. You can't calculate the core temp without also calculating the heat transfer paths. Perhaps you measured it?

It's difficult to answer your question without a lot more detail. Heat will leave the coil through many paths.

Conduction via the valve may be the dominant path, but it really depends on all the resistances in that path starting at the coil and ending at ambient air, including all interface losses. From your description it sounds like these resistances could be quite high. If so, radiation and convection become very important.

Also, don't discount convection due to "minimal airflow". Natural convection flow is typically very low (20-30 LFM).

ko (

http://www.ecooling.biz)

 

[israelkk]

Is the 204C is when a fluid flow through the valve or with no fluid? Usually when the fluid flow through the valve it cools the valve is its temperature is lower.

You can get the coil temperature from the coil resistance at stabilized position compared to the coil resistance at room temperature. When the copper coil is heated its resistance changes 4% with every 10C.

 

[btrueblood]

Dan,

Another way to answer your question would be to pull a vacuum in the oven (remove the air); the resulting heat path will be dominated by conduction, and you should see the coil temperature rise accordingly. How much rise actually occurs will give you a feel for how much of the coil heating is dissipated by convection (air) vs. how much is conducted to the body/piping.

 

[sailoday28]

btrueblood (Mechanical) Dec 3, 2004 states
Dan,
"
Another way to answer your question would be to pull a vacuum in the oven (remove the air); the resulting heat path will be dominated by conduction, and you should see the coil temperature rise accordingly."

Don't discount radiation in the oven.
How sensitive is the electrical resistance to voltage.
With or without oven, if elect resis is sensitive to temp, put varying low volt to coil, measure amps and get resistance. Temp of coil may then be determined and indirectly, the overall THERMAL resistance.
At higher voltage, one can then approx or extrapolate the coil operating temp.

 

[ko99]

Yes, the coil temperature can be found by measuring the resistance (see israelk's reply above). This method works better than a thermocouple.

However, Dan's question concerned the dominant heat transfer path. The answer is: we don't know but you can get a pretty good idea by testing.

Eliminate or modify some thermal paths to determine their effect on the results:
1. As suggested above, change the fluid flow in the valve
2. As suggested above, evacuate the oven to stops convection but not radiation or conduction.
3. You also could simply wrap the valve body in insulation to reduce convection and radiation, but allow conduction.
4. Try disconnecting the valve body from the aluminum if possible (with fluid off) to reduce conduction but allow convection & radiation. Or perhaps you can find another way to interrupt the conductive path.

Warning: Monitor carefully! These tests can increase the coil temperature, which increases heat (i2R) and could cause a runaway condition. The plastic core, solder joints, FR4, could all reach melting points.

ko (

http://www.ecooling.biz)

 

[pncdan]

The test is performed with no fluid in the valve. The valve and manifold are placed in the oven on a shelf. I allow the assembly (coil, valve and manifold) to preheat to the oven temperature 85 deg c with no power to the coil. Next I apply power to the coil and take a current reading ("I" cold). The coil is then allowed to stabilize under power. Once the current no longer changes I record this reading as my "I" hot.
I apply OHM's law to determine the resistance values that correspond to these currents. I insert the values into this formula for core temperature of coil

core temp =(234.5 + 85) rh/rc-234.5

subtracting the ambient from the core temp yeilds the temperature rise of the selinoid over ambient at this voltage

I like the vacum idea (Very smart)that would be a great way to eliminate the convection factor but I dont have that equipment in my lab.

Thanks for all the great replys
dan

 

[btrueblood]

Dan,

Do you have any kind of access to a vacuum pump/bell jar? You could do the test with the valve at ambient temperatures, both in air and evacuated, then compare to the hot air test. If all you want is a ballpark answer, this should get you close. As far as worrying about radiant heat losses, as long as the valve body temperature and oven wall (or ambient room) temperatures are the same from the air test to the vacuum test, the radiant heat losses are comparable between the two conditions.

Do, however, approach the problem "from the bottom" in vacuum test, so you don't smoke the valve; i.e. run tests starting from 1/10th or so of the usual amperage, and calculate the coil temperature before proceeding, use the data you collect to try and predict where the temperature will stabilize as you increase amperage.