RGM 920cc kit

Not open for further replies.


Jul 15, 2005

First post to list, although I have tried to look for other posting on the topic. I'm looking at the RGM 920cc kit which includes pistons/liners and looks to be way cheaper than the Norvil kit. First, I've heard some bad things about RGM. Should I be worried? Second, the kit they offer uses std size gudgen pins while norvil uses oversize. Might there be a problem with the smaller pins? Third, they claim their pistons for this kit are identical weight to the 850 pistons does anyone know if this is true?Finally, I would really like to hear from anyone using any 920cc kit-opinions, long term reliability, etc. I would like to build a runner not a grenade.

Joe Yelton
Re: your 920 questions.

Hi Joe,

I have spent the past 4 years doing numerous build-ups of a 920 Norton. My current version is a .020" over (hence the 932) kit using Norvil internals. Some of my work has been referenced by Johnny Rocket (see the 'more power' post.

I have probably spent at least 5 hrs in long distance calls with Les Emery and his cohort, Pete Lovell. Pete is a great technical guy who will give it to you 'straight-up'

OK, here is my opinion after running the 920 for 3 years. 1st, I would not use the Norvil kit again for numerous reasons. (By the way, I think the RGM spun cast liners are excellent. As far as I know, both Norvil and RGM source from the same supplier)

1) I don't like the 20mm gudgeon pin for 2 reasons: it is excessively heavy and it requires modification of the conrods and, in my opinion as an engineer it leaves the small end of the conrod a little 'light' for my liking. I am refering mainly to strength.

As an aside: does anyone out there have an accurate tensile and yield strength and material spec for Norton Conrods?? As of yet I have not found any really reliable information and I would prefer to err on the side of caution.

In order to get my piston weight close to the stock weight I had to get the gudgeon pins waisted. The gudgeon pins comprise 22% of the total piston weight. I can send you spread sheets showing the tensile stress on the small end of the conrod as a function of rpm.

2) the piston has its genesis as a Ford Galaxy piston (as reported to me by Les!!) I gather the main modification was trimming material from the piston crown and adding valve cut-outs. My main issue with the piston is getting adequate compression ratio. In order to get a CR of 9.5 to 9.7 I was forced to shave the barrel and the head. I have since made a .030 thick head gasket (water jetted) to bump the compression to 10.7 I also shortened my pushrods to keep the valve geometry original spec. I removed about .070" in total from the head and barrel. By the way I always check my CRs with a 5W oil and use a graduated cylinder to measure the volumes on the assembled engine.

As I recall the RGM 920 pistons are flat-topped and will produce a fairly low CR. Probably in the low 8's or less.

Also, with the flow work I have done with Johnny Rocket, we have come to the conclusion that Kenny Dreer was right in going to forged pistons from a reputable manufacturer like JE. The shape of the piston is not ideal for optimal 'quench' area and flow characteristics. On my next build I am going to source piston through Johnny and we will design something consistent with current state of the art automotive technology.

I am ready to order new 920 pistons for next year and, if you like I would be happly to split the cost of a custom run (usually 4 pistons minimum) from JE or a similar manufacturer.

It is interesting to note that as part of all the work I have done on my 920 engines, I have found the exhaust system produces the most dramatic effect!! Again see the posts on "more power". I have taken a bone stock 850 (actual 830cc) and installed a stepped header with an open megaphone and have noted a tremendous 'seat of the pants' increase in upper end torque (+ 4500 rpm). dyno tests will be done later this year.

My recent 920 has dyno tested at 65 HP and 60 ft-lbs rear wheel. (BTW Kenny Dreer claimed 70HP and 56 ft-lbs rear wheel). The British Manufacturers used to quote crank HP and Torque. (roughly, multiply by 1.15 to 1.2 to get equivalent Crank HP and Torque) I expect to increase the HP by at least 5 to 7 by using a stepped header design that has shown much promise in preliminary testing.

Joe, if you are not nodding off already, you may be interested to see the full range of mods I have done on my 920. (see below)

920 Engine Features:
Cylinder Head
The head is ported and fitted with oversize high-flow Valves.

Lighten Rockers (Reduces force required for rocker actuation by 20%
Pushrods have been shortened to accommodate the new valve train geometry.
High performance Norvil (racing) valve springs
Dural Retaining Collars.

Pistons and Rings and Barrels:
920 pistons are lightened by 15% and feature additional fly cutting for valve clearance on 10.7:1 compression engine.
Total seal piston rings for the ultimate in eliminating ‘blow-by’ and increased compression.
Barrels are sleeved with spun cast liners and bored to 3.215” diameter.

Set of Kehin 35mm Flat Slide Carburetors. These Carbs feature an accelerator pump that produces loads of torque when you twist your wrist. These are thorough-bred race carbs that give you strong smooth acceleration throughout the RPM range.

Exhaust: 1-3/4” Custom High Flow Exhaust System
(this has since been modified to a 1-3/8 to 1-3/4 stepped header)

Bottom End
Cams: Megacycle 56-000 Camshaft c/w radiused cam followers
Vernier Camshaft Sprocket

Crankshaft: Crankshaft is Nitrided and features a Lightened Fly wheel (Torque required for Spin up is reduced by 25%) The Crank, Conrod and Piston Assembly is balanced to the standard 52% balance factor.

Crankcase halves: Strengthened Left Hand Crankcase Half: The LH case is stiffened with a 3/8” thick 5” diameter T6 aluminum plate that is press fitted then bolted onto existing boss.

Inner primary cover is bored to fit over the beefed up LH case. The Outer Primary case has been vented for no other reason than ‘it looks cool’! It also shows off the belt drive and polished clutch spring assembly.

Conrods have been inspected for fatigue and cracks then polished.

Crankcase breather is a Harley after-market type. This results in a negative 2 psi crankcase pressure throughout the full RPM range. This means no motor-oil leaks…
GREAT post. It's nice to see a post like this that answers a lot of question for all of us that are wondering what if??

Thank you for the great information on the Norvil kit.

I too have wondered about the Norton con-rod material. I suspect its minimum yield strength is about 25,000psi, but this is strictly a guess. Perhaps destructive testing on an old rod is in order. I would love to know the results.

I don’t understand why Norvil chose to increase the size of the wrist pin? Perhaps there was excessive bearing stress on the Norvil piston bosses where the wrist pins live?

And yes, I would be interested in seeing the con-rod stress at, say, 7,000 rpm. Where do you suppose the weak link in the rod is – at the wrist pin end, crank end or in between?

I was confused after reading your statement about checking for cracks and fatigue. It’s my understanding that fatigue starts with a crack, usually at a stress riser, and propagates to failure. So, I assume you are checking for stress risers and fatigue cracks? Did you use liquid dye penetrant or some kind of UT gizmo? Am I asking too many questions?

Lastly, why did you choose polishing over shot peening in an effort to minimize fatigue induced failures of the con-rods?

Again, thanks for the excellent post.

Any answers will be greatly appreciated.

Hi Jason,

Re: con-rod material designation and metallurgical data: This info must be out there. Does anyone know of any metallurgical data? Alloy type? Heat treat?

Re: Norvil’s decision to go with the large gudgeon pins: The larger (20mm) gudgeon pin was what the original Pistons, as sourced from their supplier, came with.
There was/is a perception that the larger pin helps with heat transfer from the piston, or so Les has said. I don’t believe bending stress is an issue. Remember Kenny Dreer and his array of experts kept the stock Gudgeon pin size. In summary I think Norvil’s keeping the large pin was expedient. Again, my concerns are weight and conrod small end strength.

Re: typical con-rod failure mode: Les and Pete tell me that the vast majority of con-rod failures occur about 1/3 way down from the top. Cyclical bending stress is the culprit. Hence the often noted caution to avoid nicks and such on the fore and aft surfaces of the con-rods.

Stress at small end of conrod. At 7,000 RPM and a total piston weight of 390 grams you would have about 2,710 lbs force pulling up on the Con-Rod. If you let me know your e-mail I can send you the spread sheet. This translates to a stress in the con-rod of about 7,200 psi. The stress would be about 5,600 psi with the stock pin size.

Re: fatigue and cracks: In theory, of course, all materials subject to a cyclical load are subject to fatigue. Of course what we are concerned with is fatigue that breaches the critical material properties. You are correct in your understanding of critical fatigue. Cracks are a result of fatigue when, in turn, come from stress risers and cyclical loads. Typically a crack starts at some form of stress riser in the most highly stressed region of the part. Yes, Liquid dye penetrant is the standard. I polished the fore and aft edges of my conrods after they were inspected for cracks.

Re: shot peening: Shot peening is great if done correctly and disastrous when not done correctly. The only people I would trust is someone who has a strong history of success doing exactly Norton con-rods! If shot peening is done at too great a velocity, too long, too small diameter shot, too heavy or too light of shot, surface embrittlement or ‘excessive work hardening’ will result. At the ‘end of the day’, it appears good shot peening is a bit of a ‘black art’!

Hope this helps!

Thanks for the quick response!

The con rod stress at 7,000 rpm sounds benign to me, especially assuming a material yield strength of 25,000-psi. However, I’m not familiar with the behavior of aluminum when it’s subjected to so many stress cycles, other than it has a finite life no matter the stress level.

Moreover, my estimate on the material yield strength may be optimistic. I - like you - feel that someone out there should be able to provide the mechanical and chemical properties for Norton con-rods. My guess is that there may have been some material changes throughout the years and that some years are more desirable than others.

I agree with your “black art” description of shot peening. However, Norton shot peened the ’75 Commando rods and I often wonder if this was done in an attempt to compensate for the substitution a lesser strength material that year???

And thank you for offering the spread sheet on con rod stresses; I’d love to see it! I’ve sent you my e-mail address via PM.


More thoughts on the RGM 920 and related stuff

As I recall, the original question was an inquiry into the efficacy of the RGM 920 kit. I apologize for getting side tracked in my previous post.

The last time I talked to Roger at RGM he told me that his pistons were flat topped and, when installed, the ‘crown’ or ‘flat top’ of the piston came to more than .060” below the top of the cylinder barrel. I would be inclined to phone Roger and see if anything has changed since I talked to him several years ago. With a standard head gasket thickness of .030 and a cylinder head volume of 50cc you will get about 8.2:1 compression. I would check your head volume. This can be easily done with some 5W oil and a graduated cylinder. When you do this I recommend that you do an independent measurement at least 3 to 4 times to get a sense of your repeatability and hence accuracy. A Norton head is typically around 50 to 55cc. Again, I will email you the spread sheet to aid in your measurement and calculations.

I, personally, like to get at least 9.5:1 compression. I will have to go through my notes to find a reasonable estimate of the relationship of Compression Ratio to Torque. As I recall a 15% increase in compression will net you less that 40% of 15% in torque. (~ 6%) I will get back to you on that. As an aside, I am currently on an extended trip through the Middle East and am currently writing this from Muscat, Oman, and I do not have access to my files where I have squirreled away such gems of information.

An obvious fix to the Compression Ratio issue is to have the base of the barrels planed before the new barrel sleeves are installed. The only issue here is valve clearance. Use plasticine to verify that you have a minimum of .060” valve to piston clearance.

My next set of pistons will be some type of protruding crown design. I am going to work with Johnny Rocket on the specifics.

Back to the question at hand. The RGM kit would be OK if you don’t mind a lower Compression Ratio. Roger produces many fine Norton performance upgrades and I don’t have big concerns on his quality. (my experience only)

If I were to offer my opinion, or tell you what I would do if I were going to start over; I would buy the spun cast sleeves from RGM and have my own pistons designed.

Allow me to digress once again and tell you of my first 920 kit. As itemized in my previous post it was a Norvil kit with a big valve head conversion. I did the standard items of re-inforcing the left-hand crankcase bearing journal with a shrink fit piece of 3/8 T6 aluminum and went to a belt drive system to improve transmission life with that extra torque. I bought some 1-1/2” pipes from Kenny Dreer (his old stock) and fitted 1-1/2” std style pea-shooter silencers.

Although I had more bottom end torque, the overall experience was quite disappointing. (I will send you the Dyno charts, if you are interested.) I had nearly a ‘dead-flat’ 50 ft-lb torque curve from 3,000 rpm to 6,500 rpm. This was not the excitement that I had shelled out all those dollars for! When I changed out to a custom build 1-3/4 exhast system the performance change was dramatic!! With a straight through after market silencer on this system I gained a full 10 ft-lbs of torque at beyond 4,500 rpm. Talk about a seat-of the pants difference!!

So the moral of the story is that, even if you increase the displacement, if flow-through resonance characteristics are not favorable, you will likely be disappointed in your results. In my experience, flow-through resonance is largely dependant on your exhaust system. My advice? Before you begin your project invest in a base line dyno test. After your build up, go back to the same dyno and test again. If you do not see the results you are looking for, DON’T GIVE UP!!!!

The answer is almost certainly in flow through resonance. You can change your intake length and diameter (difficult) or change your exhaust system (relatively easy). By the way, I have made 3 different exhaust systems by buying Mandrel bends from the local auto performance shop and cutting and welding till I get the shape I’m looking for. Relatively easy to do, trust me…

Once again, I strongly suggest you invest in dyno tests. If you don’t you will always be playing a guessing game!


Thanks Brent for all the great info! My plan (and funding) for this motor is several steps below what you have done. I'm running a '74 850 with the stock 30mm ports and valve sizes (MAP valves/guides). Johnson flat track cam. Crank/bottom end will be basically stock. I'm looking at the MAP aluminum rods....

The lack of availablity of high(er) compression 850 pistons got me looking at the 920 kits but it seems that these RGM kits will still result in a bunch of machining to get the ratio up in the 9.5:1 range. The Norvil kits advertise ~9:1 but put a bunch more stress on the bottom end due to the heavier pins. Not much of a choice.

Of course I stripped the bike before even considering a dyno run.

So now I have even more questions (funny how that works). Planing of the cylinder base should be straightforward but what kind of jig did you use to flycut valve pockets in the pistons? Also is your stepped header a 2 into 1? Want to share dimensions? Finally, do you find the vernier cam sprocket is valuable (i.e. worth the money).

Thanks again for your help,

Hi Joe,

Fly cutting the pistons can be done at any good power-sports machine shop. The included angle is 60 degrees. To find the piston to valve clearance you will need to dry assemble the motor with plasticine stuck on the top of the pistons. Then turn the motor over several times with a wrench on the crank. Valve to piston clearance should be about .060”. Certainly no less that .050”

I do have a vernier cam sprocket. To install it you will need a good degree wheel. I fixed mine to the timing side of the crank using an O-ring and a golf tee. Crude but effective. Then I used a modified piece of coat-hanger wire with the one end ground to a point as a pointer. I affixed this via one of the head bolts. One of the interesting things I found was the variability in valve position owing mainly to variability in the cam as manufactured. I installed a megacycle 56-000 which, presumably, is a first rate cam.

At any rate I would suggest that you would be doing good if you get all the timings (open, close and max lift) within 3 degrees of spec. It took me pretty much a whole afternoon to set the vernier sprocket. Next question: do I think it was worth it? Probably not. Seeing the performance increases I get on stock Commandos with only alterations to the head, valves and exhaust I would say that cam timing, certainly within 5 degrees of spec, is not terribly influential on performance.

What I will do in the future is go to a crank driven ignition timing system and get away from the Boyer. There is too much flicker for my liking even in the tightest set-up.

With respect to piston choices, phone Les at Norvil to discuss. Norvil may offer some alternatives now. Les always has some worthwhile insights and entertaining stories. They claim to never have had an issue with their 920 kit.

As I had mentioned earlier, I am going to have some pistons made. Apparently they are done in a minimum quantity of 4 so it is entirely likely I will have two left over. I will keep you informed of my progress here if you like.
Not open for further replies.