Urgent: How can I stop bolts from continually coming loose

Not what you want to hear, ... But weld the suckers, then it'll be a problem of how to undo them in the future;)

There are some special segrated washers available that I have seen on a few different things that have a habit of coming loose .Konnect / Coventry Fastners have them from memory.

Frank, are the bolts stretching? if this is the case, what you need is harder bolts. Marking the heads so you can tell if they turn may help you to be sure whether they are stretching. If they loosen by initially stretching and then turning, retightening them will see them tighten past the previous marks.

John

One more option is to go close tolerance bolts.

Looks like you've had plenty of suggestions tank, something must work.

Thanks once again for some very good tips/advice.
I am convinced that due to the weight of the compressor, which is at least 15kgs. and it is driven by a rubber coupling shaft from the back of the injector pump that the clamping force of the 4 bolts is not enough to support the weight, I believe that 2 solid dowel pins to locate the bracket and support the weight would allow the bolts to clamp and hold tight.
I think if I fit 4 close fit studs with the shanks supporting the weight the 4 nuts will have a chance at staying tight, I will give it a go tomorrow and report back the results after the next trip north, John (JDNSW), the bolts that came loose were new grade 8 bolts which I measured with a micrometer before installing, they are still the same length, the threads in the block seem to be good the new bolts were fitted without the bracket, hand tight then lightly nipped, no play or movement, the shanks on the old bolts were smaller dia. than the thread as are the last new bolts fitted, will try to get new bolts with thicker shanks as I believe that the shanks will support the weight, like a couple of dowels would, Regards frank.

4x 3/8th bolts should be able to support in the order of tens to hundreds of thousands of pounds of load.

if you're checking thread condition by seating the bolts and then tighteneing them and then checking for free play you're not effectively checking the condition of the threads in the block. I suspect that the root cause of your issue is the bolts and the way you are putting them in.

try this.

Use a high temp loctite, 292 or one of the bearing mounting locktites. Pre heat the block and the bolts, get the block to about 60 degrees and the bolts to around 80 degrees. Use grade 10 bolts (just confirm that your grade 8 unfs had the 6 radiating lines in line with the Apexs of the head flats and not the center of the flats.)

I think that studs and locknuts with locating dowels should be all that is required Frank. Go up to7/16" UNC if possible or even 1/2"UNC. Can you post a pic of the assembly bolted up?

JC

G'day Tank,
Hang in there and we all are waiting for the final solution.
Just a thought. Are the bolts bottoming out in the holes and the bolts are to long?. When you torque them up they maybe bottoming out slightly before it can pull down the plate to the correct torque.
Chris

Frank it was mentioned earlier but this diagram might better explain, being a Perkins might the threads in the block be whitworth form?

http://www.aulro.com/afvb/attachment...1&d=1458164300

If not, then I think you are on the right track with dowelling it.
However, from your description, it sounds like a stock installation (?) so it doesn't make sense that modification is necessary to make it stay there.

Chris, JC and Dave, First and second time I used a Loctite product ($35 for the smallest bottle they could find can't recall the product # but it was for high vbration areas on jet aircraft) and the third time, each time the block threads were cleaned with a plug tap screwed in fully by hand (no metal removed), then Shellite, then a large bore rifle barrel wire and fibre brush's.
The bolts were similarly cleaned, then screwed in by hand and checked for play on the way in, then nipped up by hand, no play.
This bracket is a cast iron affair with a rectangular flange on the block side, it curves up to be the bottom sump plate of the compressor probably 8"x4" with multiple small bolts to seal and attach to the crankcase of the comp..
This bracket is hollow, as it is the drainback to the engine sump from the comp..
It is in a very awkward place and only 2 of the bolts are easily accessible, I had to buy a special tool to be able to torque one of the bolts, I am using a near new W&B torque wrench (one I had for over 30 years was pinched).
I have spoken to an engineer at Perkins and I suggested either Dowels, which would require the engine out as there is no room to drill dowel holes in block, said bolts with small clearance shanks would be as good.
There are many problems with supporting brackets, as they may make vibrations worse, I have a similar problem with the oil/water heat exchanger which has the 2 oil filters attached cracking the alloy housing, it would weigh at least 20ks. and it all hangs off 5 long and one short bolt.
As the engine is turbo'd it needs the extra oil cooling, thinking of fitting a remote oil cooler (aftermarket) with a dedicated fan, sick and tired of oil leaks.
I seem to remember years ago when I was fitting a Road Ranger box to a Perkins 6354 that perkins had a kit to fit 2 dowels the the flywheel housing, so it seems they didn't learn and put some dowels in these brackets that carry so much weight out and away from the block, will get some pics and let you know how it turns out, Thanks again, to all. Regards Frank.