150 to 210 is the range that my box shifts smoothest. That said there are a lot of factors into temps for any given car. Driving conditions, lube type, LSD or not, and driving style all contribute. I used an infrared thermometer for a good while but was never satisfied that I got good readings since the ground clearance limited the number of places I could get a good shot at it. In other words, there was too much variation in the readings to trust them. My temps with the installed sensor/gauge are lower than those I got with the infrared thing.
I figure the 993 engine driven like I stole it puts enough stress on the R&P that I want as slippery and cool a trans fluid as I can build into the car. The excellent clutch-type Guard LSD just furthered the cause.
Also I installed a temp sensor (shown in the thread below) in only one place, where the speedo gauge would go. That works for me because I use GPS as my speedo and don't even put a speedo gear in my transaxles any more.
Next time I take the box out I'm going to put sensors in each compartment just to see if there are significant differences.
http://www.914world.com/bbs2/index.php?sho...161520&st=0Chances are you're fine without a cooler, even at 200 degrees if your box shifts well. I'm also convinced that there is a lot of exhaust-heat soak if the car is in town sitting at red lights and hardly moving... not much of a factor on the open road. I run synthetic fluid in mine.
Initially I used it because, with a high torque engine driven with authority, I would be protecting the R&P and they're more expensive than synchros. But I also discovered (although this is prior to having a temp gauge) two things.
1. My temps dropped about 10 degrees on average and
2. My synchro wear didn't increase at all. About that ...
Number 2 is conditional: There are plenty of folks that will tell you synchro's will wear badly and dog teeth, because the synchros don't hold as well, will suffer if you run synthetic. They might be right for most folks.
All I can tell you is that I haven't found it to be the case. But I'm certain it has everything to do with shift technique, i.e. smoothness and reasonably "slow" shifts, but I can't prove it. Not enough data points.
Good luck