Martin, if you have access to both ends can you pull the bush in using a suitable piece of studding and some clamps? you may need to make a guide or stepped washers to keep thins square.

Chris Gunn

How much interference? 1 thou per inch* is 100°C with steel, 3 thou per inch is ambitious. Both should allow a slip fit with heating and cooling. More than that is generally unproductive.

*Other measurement systems are available e.g. 1mm/m; 5' 3"/mile etc.

Don't even bother putting the bush in the freezer. Temp difference from room temp of 25C to freezer of -5C makes little difference to the size on a small item like this. But the bigger the temperature differential between the bush and the housing, the faster the heat will rush out of the hot object into the cooler object, thus shrinking the former and expanding the latter.

If you make your bush with the usual one thou per inch of interference fit and then heat the housing to 150 to 200C with a propane torch, it should just about fall in. (The amount of expansion of steel being roughly one thou per inch per 100C.) And you can put hi-temp loctite on it too if you feel you must.

Being prepared with a suitable stepped drift and large hammer, plus an even chamfer on the leading edge of the bush (and the hole it goes into) to guide it in would be good preparation. Or stepped plug with a large nut and bolt through it to a flat plate on t'other side of the housing would pull it in nice and even too.  For something the size of your job, I use a 9/16" UNF grade 8 bolt and nut. Those fine threads will exert A LOT of force with a large ring spanner or socket and breaker bar on the nut, with plenty of anti-seize on the threads.

If movement is a critical concern — and once again we have been given no idea of the application for the parts in question so we can only guess — you can always drill and dowel or drill and screw one or more places around the OD of the bushing to mechanically lock it in position. Very rare that this needs done in this age of Loctite though.

Edited By Hopper on 28/03/2018 05:57:13

If it were me doing this i would use a bigger press – if you have ever had a bush freeze half way in no manner of hammering will move it . I wouldn't bother with retaining compound as a properly designed press fit won't move once it is in and with some bushes you have to use an anti galling compound on the surfaces , you could champfer the bush and stake it in if you really want to .

Edited By XD 351 on 28/03/2018 07:03:24

Tunnel vision here, I reckon. If liquid nitrogen is available, it is obviously the best way to install interference fitting parts. Warming the housing would likewise increase the clearance for fitting.

But, without being provided with all the details it is not clear of the loading on the bush in use, etc. How long is the bush? What is the ID? Axial or radial loading? Plain bush or lipped? Through bore or blind? Will the bore be machined after installation?

Quite likely a slip fit with a suitable grade of loctite, or similar, would suffice. Quite likely a couple of grubscrews, on the joint line, would add further security.

Personally, I don’t see any particular problem given the scanty information provided.

Obviously it will work with that large interference, but 250C will have seen off your loctite, The difference in using liquid nitrogen rather that heat is in radiant heating150C to 15C gives (150+273)^4-(15+273)^4 = 2.5e10 to drive the rdaiant heat, but 15C to -120C (same difference) gives (15+273)^4 – (-120+273)^4 = 6.3 e9, only 1/4 of the radiant transfer.

If it is a clutch pulley from an ML7 and the sole purpose of the bushing is to allow the pulley to spin on the shaft under no load when freewheeling in the disengaged position, heavy press fit plus heat plus Loctite was way overkill. Light press fit (one thou total interference) or even a push fit with Loctite would have done the job.

Sorry, I rather thought it did after reading your post: "Once brass bush from my ML 7 pulley fitted in old Myford (Nottingham) gone loosie I have decided to add Loctite as an additional insurance in power transmission systems whenever practical."

I must have misunderstood.

So we still have no idea what this Top Secret pulley's application is. Whether it will be transmitting the power from a 50hp motor at 1400rpm or acting as a belt tensioning idler on a 2rpm hand cranked glass display cabinet, or one of the 10 million possibilities in between.

Which means there is no way for anyone to know whether they are offering you relevant advice or not, as specs and methods vary accordingly. It leaves everyone guessing in the dark and a good chance you end up with duff advice.

I'm not sure why we get so many questions about this type of Top Secret project where no specific details can be given about the application or the exact job in hand. Must be a lot of mad inventors on here trying to keep one step ahead of the competition.

Edited By Hopper on 29/03/2018 08:25:42

Dr Stangelove and the 200k centrifuge eh.

Ah well, if pulley is used for driving a grinding wheel, a good crush fit is probably appropriate tp ensure no movement. How will you attach pulley to shaft? By cutting a key way int he cast iron bush?

Yes, shrinkage like that of the ID is pretty common with a heavy press fit. Metal is not as immovable as we imagine it to be! Another way of doing it is to make the bush ID undersize, press it in, then set the whole shebang on the faceplate and bore the bush to final size in place, thus ensuring it is perfectly concentric with the OD.

Ah yes, the set screw. You might need a long series drill bit, or drill the hole in at an angle. I don't suppose the key will know any different if the screw is angled a bit.