Is a Heat Exchanger Necessary for Optimal Gearbox Design?

[Teplorod]

I need to make thermal analysis for new gear box design to get the answer: do I need additional heat exchanger for that gearbox?
I would like to get a help as a source for engineering calculations. Of course, I have already calculated power losses in bearings with TIMKEN software. I don’t have enough information how to estimate oil churning losses and other power losses in gearbox. And finally, what the best modern calculations methodology you can suggest me for temperature balance?
Thanks.

 

[CFDFEAGURU]

How much heat is the gear box generating? What is the oil temperature at the outlet? What temperature do you need to maintain the oil at?

Thanks
Matt

 

[Teplorod]

Thanks for response Matt,
This is my question, how much heat gearbox is generating? I need estimate this on the design stage. I understand that generally there are power losses at tooth engagement, churning power losses, bearing and seal power losses. How to calculate them?
Thanks.
Vlad.

 

[FredGarvin]

If it were me, I would estimate the losses at around 15%. Assume whatever power you are putting through it, that percentage is put directly into heat and design accordingly. Since you are at the design stage that would at least get you going.

 

[Q_Goest]

Break down the heat input into the various parts that create heat:
- Bearings (rolling element or hydrodynamic)
- Gears
- Windage
- Other frictional losses?
- Oil pump losses

Rolling element bearings have a very low coefficient of friction, around 0.002 or less. Check with SKF or other bearing manufacturer. Hydrodynamic bearings are similarly very low. Use that frictional load at the point on the radius where the bearing is and use the radial load as the normal load to calculate the frictional load. Then determine heat input by multiplying by velocity to get power.
http://www.roymech.co.uk/Useful_Tables/Tribology/Bearing Friction.html
http://www.roymech.co.uk/Useful_Tables/Tribology/Plain_Bearing Friction.html
http://www.bearings.machinedesign.com/guiEdits/Content/BDE_6_1/bdemech6_3.aspx
http://www.skf.com/portal/skf/home/products?newlink=1_0_40&lang=en&maincatalogue=1

Gears can be looked at simply as a 'loss'. Gears are x% efficient at transmitting power, so that x% gets converted to heat that must be rejected.
http://www.roymech.co.uk/Useful_Tables/Drive/Gear_Efficiency.html

Windage probably isn't very significant, and is very difficult to analyze with any accuracy. Just make sure the rest of your calculations are conservative and you'll be fine.

If there are any additional frictional losses such as a sliding bearing (ex: linear or cylindrical bushing) then you can calculate losses by estimating frictional coefficient times normal load times velocity.
http://www.roymech.co.uk/Useful_Tables/Tribology/co_of_frict.htm

An oil pump is essentially a direct heat input. If you have oil being circulated into the box using a pump that requires x amount of power, you can simply add that amount of power directly to the gearbox. That power has to come back out somehow, which is why an oil cooler is often used.

Once you have all the power losses, determine the steady state heat transfer out of the box by determining how warm the box has to get in order for that heat to be rejected.

 

[Bob S]

The 5-speed manual transmission in my car is supposed to be ~96% efficient in all but 4th gear (1:1) which is ~98%. Unlike automatics, it doesn't need water cooling.

Bob S