Gearbox oil cooling
Discussion
I've had several different failures over the years with the saab gearbox , blowing diffs and stripping gears but the last three have been synchro failure on 2nd gear
It's gone to a specialist this time as the man I use locally doesn't do this sort of work .
The specialist is talking about adding a pump and cooler as he thinks it's a temp issue with the oil not being able to lubricate the synchro properly , .
The last three failures have been the same and always on trackdays .
I've been reading what I can find on the net which isn't much regarding temperature and I have been using a Morris's lodexol mtf for the last few years .
does anyone know how hot the oil can get in a manual box and what does it do to the oil that stops the synchros from being able to grip the cone on the gear . Tia
It's gone to a specialist this time as the man I use locally doesn't do this sort of work .
The specialist is talking about adding a pump and cooler as he thinks it's a temp issue with the oil not being able to lubricate the synchro properly , .
The last three failures have been the same and always on trackdays .
I've been reading what I can find on the net which isn't much regarding temperature and I have been using a Morris's lodexol mtf for the last few years .
does anyone know how hot the oil can get in a manual box and what does it do to the oil that stops the synchros from being able to grip the cone on the gear . Tia
gearboxes cause losses of about 15% of the applied power through the gearbox. Some less, some more.
So you dissipate some 10-12kW of heat in the gearbox for every 100HP your engine generates.
Given a lot of power and a small uncooled GB, you can easily see 150C of oil temps in a GB.
That is why race cars often use high viscosity GB oils and oil coolers if required.
Such gearboxes are also often equiped with integrated oil pumps, so you just need to hook up a suitable cooler or oil/water heat exchanger.
As the oil gets hot, its viscosity drops. The oil film thickness under pressure also drops. Or the other way round, with increasing temperatures, the maximum forces you can transfer though a oil film without the metal surfaces touching each other drops.
This is a simplified view. There are also other destructive phenomenons taking place when the oil film is just too thin.
You might do 3 things:
1. try to load the syncros less, e.g. rev match on down shift, heel and toe, and improve your upshifting timing. if you bang in all the gears you use the syncros as a clutch to rev match the engine to the gearbox. The syncros should only do the fine matching. The bigger part should be done by the driver, especially under hard use.
2. fill the GB with a heavier oil, e.g. 75E90 or 120. Cold shifting masy be impaired at some point.
3. fit a gearbox cooler or heat exchanger (plus pump) and consider 1 & 2.
So you dissipate some 10-12kW of heat in the gearbox for every 100HP your engine generates.
Given a lot of power and a small uncooled GB, you can easily see 150C of oil temps in a GB.
That is why race cars often use high viscosity GB oils and oil coolers if required.
Such gearboxes are also often equiped with integrated oil pumps, so you just need to hook up a suitable cooler or oil/water heat exchanger.
As the oil gets hot, its viscosity drops. The oil film thickness under pressure also drops. Or the other way round, with increasing temperatures, the maximum forces you can transfer though a oil film without the metal surfaces touching each other drops.
This is a simplified view. There are also other destructive phenomenons taking place when the oil film is just too thin.
You might do 3 things:
1. try to load the syncros less, e.g. rev match on down shift, heel and toe, and improve your upshifting timing. if you bang in all the gears you use the syncros as a clutch to rev match the engine to the gearbox. The syncros should only do the fine matching. The bigger part should be done by the driver, especially under hard use.
2. fill the GB with a heavier oil, e.g. 75E90 or 120. Cold shifting masy be impaired at some point.
3. fit a gearbox cooler or heat exchanger (plus pump) and consider 1 & 2.
Edited by Ive on Sunday 15th February 11:49
Until you do some tests you won't know for sure so that's where you need to start. I once pondered if it was going to be an issue so stuck some heat strips on a box, they showed that at least the outer casing wasn't going over 100'c so I figured it was ok - an internal probe would be a better idea though.
There is some Gbox oil out there called 'Joe Gibbs 75/110', that maybe what you need as well as a pump and cooler perhaps....
There is some Gbox oil out there called 'Joe Gibbs 75/110', that maybe what you need as well as a pump and cooler perhaps....
I added gearbox cooling in my noble for this very reason, what I actually do is turn it on at track days or when using the car really hard... simply don't need it at other times.
I do have a temp input into my ecu an output to turn the pump on if i choose to get more cleaver about it but I simply have not bothered yet... people who have monitored the temps have seem in excess of 120deg c on track days so its definately a must for our high power cars... you can build in a little thermostat to turn it on at say 10deg c.
I do have a temp input into my ecu an output to turn the pump on if i choose to get more cleaver about it but I simply have not bothered yet... people who have monitored the temps have seem in excess of 120deg c on track days so its definately a must for our high power cars... you can build in a little thermostat to turn it on at say 10deg c.
Ive said:
gearboxes cause losses of about 15% of the applied power through the gearbox. Some less, some more.
Nowhere near that. 15% is a good approximation of the total losses from crankshaft to tyre contact patch but only about half or less of that is in the gearbox. The rest is in driveshafts, diff and the tyre itself.I've gathered some bits together to turn it on and off based on temp and switch a small 8" fan ducted onto the cooler with the outlet under the car facing backwards . Discussing it with other Saab guys running similar power I do tend to run longer sessions on track and have had more failures than a mate with an identical motor and box set up .
I don't know wether having to cool and lubricate the final drive makes it harder work for a fwd box but the box builder has done cooler systems for transverse noble v6
I don't know wether having to cool and lubricate the final drive makes it harder work for a fwd box but the box builder has done cooler systems for transverse noble v6
Pumaracing said:
Nowhere near that. 15% is a good approximation of the total losses from crankshaft to tyre contact patch but only about half or less of that is in the gearbox. The rest is in driveshafts, diff and the tyre itself.
You're right. it is less for GB, especially with separate diffs. So make it 5-10% for GB with integrated diffs and a tad less for one with a separate diff.Schäffler shows some numbers on page 2:
http://www.schaeffler.com/remotemedien/media/_shar...
in 5th gear losses are as litte as 4% while in 1st they are 40% for this particular example. Likely not the most modern box, though.
This is a little off topic and not to do with cooling so I apologise.
Thanks for that post Ive, my German isn't up to much.
It lead me on to a case study in power losses in helical gear mesh by Petra Heingartner and David Mba. I attach a pic of the results graphically and tabular. Overall loss low and very little increase in loss with moving from 2238Kw input to 8952Kw input. Power loss went from 106.77Kw to 125.33Kw.
Peter
Thanks for that post Ive, my German isn't up to much.
It lead me on to a case study in power losses in helical gear mesh by Petra Heingartner and David Mba. I attach a pic of the results graphically and tabular. Overall loss low and very little increase in loss with moving from 2238Kw input to 8952Kw input. Power loss went from 106.77Kw to 125.33Kw.
Peter
High oil temp failures in transmissons generally come from two main sources:
1) Casing expansion (generally alluminium) leading to gear misalignment and hence incorrect tooth root loads -> tooth failure
2) Break down of oil film thickness in bearing elements under high non axial loadings - > bearing failure
Although colder oil is thicker and more viscous and will allow the syncrocones to provide a higher drag force during their engagement, its not a big factor once oil temps climb above about 50degC. Chances are, most syncro failures are due to the the engines inertia or torque output during shifting leading to overload (big issue on a turbo engine with a plenum full of high pressure air, when closing the throttle takes some time to reduce flywheel torque to a negative value!)
1) Casing expansion (generally alluminium) leading to gear misalignment and hence incorrect tooth root loads -> tooth failure
2) Break down of oil film thickness in bearing elements under high non axial loadings - > bearing failure
Although colder oil is thicker and more viscous and will allow the syncrocones to provide a higher drag force during their engagement, its not a big factor once oil temps climb above about 50degC. Chances are, most syncro failures are due to the the engines inertia or torque output during shifting leading to overload (big issue on a turbo engine with a plenum full of high pressure air, when closing the throttle takes some time to reduce flywheel torque to a negative value!)
ive blown diffs before fitting the Atb and since then it's been the same failure each time . The rest of the box looks good as far as gear wear etc . We all run the same clutch set up for 450+ and we put an adjustable push rod in the master to stop any drag at high rpm . Flywheels are lightened but not super light . Had a look at some Gopro stuff from cadwell last November and I was out for nearly 40mins in the afternoon when the synchro went
If you have a "Logging ecu" set it to record as fast as possible, and look at engine rpm, plenum pressure and ignition angle during a shift event. You probably find you are not just fighting engine inertia but positive flywheel torque as well. I've had some success with an ignition cut on a non-sequential gearbox where it detects a sudden lift of the throttle and cuts ignition completely for a time period (say 250ms) to ensure negative flywheel torque during upshifts.
So what sort of oil temps are good or bad when it comes to gearboxes ?
The little VDO pump looks decent, not sure how tolerant of heat it would be, seems to be an OEM BMW/Audi used unit so should be reliable on that front.
The cheap water pump mentioned is popular in the US as an oil scavenge pump....except they seem unreliable and must be mounted upright. people also complain about them being very noisy
For a quiet reliable oil pump that can handle heat and be mounted in any position, the Turbowerx Exa is the one to have. Not cheap though
As for shifts...what sort of clutch ? I'd guess a big heavy friction disc setup wouldnt be helping matters either ? And are you shifting at much higher rpm than normal ?
The little VDO pump looks decent, not sure how tolerant of heat it would be, seems to be an OEM BMW/Audi used unit so should be reliable on that front.
The cheap water pump mentioned is popular in the US as an oil scavenge pump....except they seem unreliable and must be mounted upright. people also complain about them being very noisy
For a quiet reliable oil pump that can handle heat and be mounted in any position, the Turbowerx Exa is the one to have. Not cheap though
As for shifts...what sort of clutch ? I'd guess a big heavy friction disc setup wouldnt be helping matters either ? And are you shifting at much higher rpm than normal ?
It's partly me and going over to a ball bearing turbo has put the box under more load lower down the rpm range .
I normally don't need to go past 6500 on track with the garret .
There is a circlip mod on 3rd gear that needs to be done as it was a factory mod on very late saab 9000 boxes and std on all the later 9-3 9-5 . The earlier boxes can have some movement under high load which is another factor with mine .
I normally don't need to go past 6500 on track with the garret .
There is a circlip mod on 3rd gear that needs to be done as it was a factory mod on very late saab 9000 boxes and std on all the later 9-3 9-5 . The earlier boxes can have some movement under high load which is another factor with mine .
mighty kitten said:
Conventional 10" sprung plate with a Sachs race engineering cover which will hold 450lb ft reliably . I can data log quite a bit of stuff from the Saab trionic Ecu but I doubt there is anything I can do to cut ign
I'm talking about the inertia the frictions discs will carry at higher rpm's, where large diameter will be more hurtful.This is also what the synchros need to slow down when shifting, as well as the rest of the gearset
Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff