Paul The MV89 is a double oven unit, and assuming it is setup and working correctly, it's sensitivity to thermo capacity, external temperature effects, sweep spot, and all the other nutty stuff, is Way below the MV89's noise level which is generally under 2e-12 when just setting on a bench in the open. (as long as you do not move or tilt it) All the stuff that is being talked about can effect single oven units. The main purpose of the outer dual oven units, is to isolate the freq change from these effects, and on the several MV89's I tested, it's outer oven works very good and the unit does not need any special mechanical considerations to improve it's stability.
Your test data, that shows such high ADEV values is most likely your tester's resolution, assuming you have a clean electrical setup. If your tester does not have at least 1ps (1e-12 seconds) resolution, which few do, then your tester is limiting the ADEV at 1 second. >From your data, it looks like your tester is very likely the problem up to at >least a couple hundred seconds. In summery, looks like you need a better tester if you want to do any valid performance testing below ~1000 seconds on a typical MV89. ws *************************************** Paul A. Cianciolo paulc at net.net Tue Jan 28 22:53:38 EST 2014 Charles, Thank you for the information and I apologize for responding late. I understand the concept of the "sweet spot". I made the changes you mentioned isolate the OCXO thermally from the chasiss. It is out of the Dewar and sitting on a piece Styrofoam. It is then mounted in a chassis that 19" wide 19" deep and 5" or so high. Sure enough after I made the changes the, ADev on time lab shows at 2 sec. all the way up to 3600 sec. With a sample rate of Hz dictated by the 1 PPS out of the thunder bolt the reading are below 2 sec. 1.79 -10, 20 sec 1.95 -11 200 sec 6.83 - 12 1000 sec 1.70 --12 Can you mention the size of the aluminum box you are using to add thermal capacitance? I am very interested to try and improve my results Thank you Paul A. Cianciolo W1VLF _______________________________________________ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.