Fullauto head before and after comparisons (2014)

"So the question should be does good results on a low velocity flow bench really mean a performance improvement ?" It depends what you mean by low velocity. It's better to test at 28"w than 10"w because flow problems that do not show up at lower velocities can be there at high velocity. The highest velocities happen at low valve lift when vacuum in the cyl is highest so testing those important low lift areas at higher vacuum isn't a bad idea. Testing on a flow bench is a great asset, it provides a measure of the exact thing you are pursuing porting the cyl head in an effort to obtain a better performing engine. Port shape is crucial to improving flow. Everyone is fairly aware that making the port bigger to get high flow numbers can be detrimental to performance, as port velocity is important. So a flow bench can allow you to experiment with shape and size against flow values. Something I've been playing with.

At the moment I have a BSA Thunderbolt head ported to use twin 34mm carbs. A BSA Lightning twin carb head has 30mm carbs and a poor inlet port shape. I measured the flow on one yesterday with a bell on the port, it flowed 112.6cfm @ 28"w. With the 30mm Amal attached without a needle(couldn't find one) it flowed 100.9cfm. The 34mm ports on the Thunderbolt head are bigger, but also drastically different in shape plus the valve is 1.5mm oversize. This will possibly drop air velocity at lower rpm except there is a point where the additional flow becomes fare greater in proportion to the additional size which means there will be a significant increase in air speed through the port, where that cross over occurs may be detrimental at some point of lower rpm, but the extra flow and speed is going to extend a useful rpm range. Until I actually test it I can only guess at the actual result but the flow bench allows me to work the head and measure so I have data to work from.

At .200" of valve lift I have 115cfm through the 34mm carbs, at .350" I have about 160cfm through the carb (170 through the port and manifold with a bell.) So a 60% increase in flow through the carb with a possible loss in port velocity in a rev range that could be below 3,000rpm where the engine is hardly used and is inconsequential. Though I need to test that theory on an engine. It could be that running the engine at 750cc is more suitable or increasing compression is needed to increase intake port velocity to add power lower down. What I know, with fair certainty, is it will have an extended usable power range developing more power.

Cool stuff there Mark.

acotrel said:

' To Alan's point, of course if you ultimately flow more air through the same port, this will support higher rpm operation, but it is not necessary to explore this high end operation regime because the engine will make plenty of power throughout.'

When you fit E3134 race cams to an early 650cc Triumph Thunderbird you get an increase in power right through the whole rev range, even below the cam spot. If you port the head bigger you get more flow however you lose midrange torque, so tend to rev the motor higher. The better option is to keep the inlet port size standard and raise the overall gearing. Most of our old 650 Triumphs have blown up their bottom ends and not returned to racing. Standard Triumph valve gear lightened and polished, with the right springs, will cop 10,500 RPM continually for about 4 race meetings a year for a couple of years. The torque thing is very deceptive, and unless you've experienced it on a high geared race bike, it is difficult to recognise when you have it. I often wonder about torque readings off a dyno. I think most guys look for the peak power reading and don't recognise the rev limit imposed by the bottom end when you race the bike.
What have you got when you raise the overall gearing and the bike accelerates faster ? It is not what you would expect. If you've got all the gears very close, and the overall gearing very high - with lots of torque it is a real thrill, and long as you don't need a low first gear.

Click to expand...

What seems to be the fly in the ointment is just making ports bigger makes a proportional increase in flow, and effects gas speed in direct relation to that. In practice a bigger port of a better shape can result in so much additional flow that gas speed is also increased in the larger more efficient port and is higher than that in the smaller slower port. That has an interesting and profound effect on power as more charge is carried in the higher speed bigger port forcing into the cylinder as the valve is about to close.

In use a stock compression stock cam A65 650 pulls much harder when the throttle is opened at any speed and naturally the rpm ceiling is raised significantly but doesn't need using. Riding this stock Firebird with the ported head could be described as 'nice', it has brilliant response at any rpm and when opened onto the main jets it goes nuts.

A 745cc A65 outfit, which was the reason I experimented with this head, dominated the Queensland P3 sidecar championship last year with a similar Thunderbolt head running on petrol, it was against bikes including 2 hot and very lovely, 1200cc Vincents. It never exceeded 6,800rpm though it surely could have, and it lowered the lap record by 4.4seconds in the process. It's ports are also 34mm but uses a Weber carb.

Update today, Matt racing the outfit at the weekend; 6 races 5 wins and a second. With a problem of it jumping out of 3rd. Nice looking Triumphs and Nortons and Vincents.

Just for reference if a R3 had three Thunderbolt stock ports flowing about 108cfm each, like the 27mm port on the right, in a hp to flow prediction it should make 77hp from a stock 750. And they had the Thunderbolt head in 1962? Though it was a BSA not Triumph, so the designers probably didn't bother looking.

So for anyone interested I started porting the std 30mm 109 head, I hacked out the bowl area and got it to 122. I tried a little bell on the 30mm entry which didn't make much difference before, but did with the bowl better, 130. So I egged out the 30mm carb end for an oval to 34mm round manifold. It 's 30mm high and now wider. That got me 135 bare and 139 with the bell all @ .388" valve lift. Interestingly at around .400" it's 142 and @ .489" its 147cfm. But that won't do any good, I don't want to use high lift, I need it flowing at stock lift. Hi lift would wear stuff out and limit rpm.

But this is a bit unusual as the stock head had around 109 after .300" and more lift made no difference. I've put some JBweld on the floor and filled it by about the same amount I took out. So tomorrow when it's cured I'll shape it and try again. What I'm aiming for is a lot more, but I need a 42mm valve and open it a bit more in the bowl and everywhere else. As I don't have a 42mm I'll see what the std one will do.

Hello All,
I re- angeled Yves Fullauto- head on my fixture and found, that his Fullauto- head was of much better quality than the original Commando heads. The job was, just re- angling and installing the bigger valve seats and no port- work, but that's beside the point. Here is the main difference:
My fixture is designed in a way that by changing from right- hand to left- hand we have absolutely the mirror- image being 25° to the center line of the engine. So one cumpustion chamber is 25° to the left and the other one is 25° to the right.
When I changed over frome left to right, the positions of the valves in the y- axis of my bore/ milling machine are always 0,5-0,8mm apart with the original heads. I don't exactly know the reason why that is, but I suspect that, within the manufacture of the cylinder heads in the old days they must have needed to change from one to another fixture. That's why we have to big diffeneces.
Another word about flow- figures:
Certainly the ports should flow as good as possible, but the flow- figure is not all. If you look into an inlet port of a fully assembled cylinder- head at TDC, you already see the open exhaust valve as well. This shows that we have an open door for the fresh gasses to directly pass into the exhaust port. This robs a lot of gas which otherwise would be in the combustion chamber for burning and producing power.
That's why Paul Dustall recommended strongly his pocket above the inlet valve seat. I guess, that this pocket will destroy a lot of impressive flow- figures, but is does lead the gas stream into the cylinder- bore of the engine. So, besides having good flow- figures we also must make sure that the gasses stay inside the cylinder- bore and not disappear through the exhaust port. By the way, Graham Bell wrote the same thing in his book "Performance Tuning in Theory & Practise".
Klaus Monning

Klaus Monning said:

Hello All,
I re- angeled Yves Fullauto- head on my fixture and found, that his Fullauto- head was of much better quality than the original Commando heads. The job was, just re- angling and installing the bigger valve seats and no port- work, but that's beside the point. Here is the main difference:
My fixture is designed in a way that by changing from right- hand to left- hand we have absolutely the mirror- image being 25° to the center line of the engine. So one cumpustion chamber is 25° to the left and the other one is 25° to the right.
When I changed over frome left to right, the positions of the valves in the y- axis of my bore/ milling machine are always 0,5-0,8mm apart with the original heads. I don't exactly know the reason why that is, but I suspect that, within the manufacture of the cylinder heads in the old days they must have needed to change from one to another fixture. That's why we have to big diffeneces.
Another word about flow- figures:
Certainly the ports should flow as good as possible, but the flow- figure is not all. If you look into an inlet port of a fully assembled cylinder- head at TDC, you already see the open exhaust valve as well. This shows that we have an open door for the fresh gasses to directly pass into the exhaust port. This robs a lot of gas which otherwise would be in the combustion chamber for burning and producing power.
That's why Paul Dustall recommended strongly his pocket above the inlet valve seat. I guess, that this pocket will destroy a lot of impressive flow- figures, but is does lead the gas stream into the cylinder- bore of the engine. So, besides having good flow- figures we also must make sure that the gasses stay inside the cylinder- bore and not disappear through the exhaust port. By the way, Graham Bell wrote the same thing in his book "Performance Tuning in Theory & Practise".
Klaus Monning

Click to expand...

Yes on the BSA the bowl is deeper. With the idea of turning charge down but I make the top of the port one smooth turn, an S bend reduces flow. On the std BSA the charge goes across the valve and through the top of it aimed straight out the exhaust, the short side of the valve doesn't do much, why I turn that smooth charge down as squarely onto the back of the valve as I can, that's when the flow numbers jump up. A 41mm valve and 30mm port should not flow only 109cfm. I have noticed a std A65 at 9-1 can ping fairly easily. Yet with the modified head there is no sigh of it, and it may be because of how mixture enters the chamber. The difference on the road is you would not even consider needing a 750kit.

I once had a 1969 Lightning, & what ever I tried I could never completely eliminate the pinking. Long after selling it I started to think that it had been the Boyer ignition which had caused the problem. Maybe it wasn't.

The std valve 40.5mm, I think it actually is, will do 160cfm according to my vacuum cleaners, it's no doubt a bit wrong but the difference to a stock port is light years. The valves are a bit low on the seats so I'll do them 42mm like the Firebird and try and get the rest. And I have another good head with good standard seat heights and I might try it with the good stock valves. I doubt there is any advantage but I already have them.

Some of the video of Matt's sidecar shows what fun that must be, he is not getting past a nice Triumph twin very easily but has the legs on it by the look, same with a particularly fast Norton, and I'm thinking he has got 745cc though it's spinning up a fair bit. I asked my friend Nick about the other outfits, they are both on Methanol like the 1200 Vincent that's so lovely. Apparently the 4 speed is at a disadvantage off the line so he has to run them down, and a 5 speed would be nice. My guess is some of the others like the Triumph may have one.

So the fast Triumph is 840cc the Norton is 920cc. And so far the 745 on petrol has taken 8 firsts and a 2nd from his last 10 races, and they took last years Period 3 Queensland championship. Australian titles are toward the end of the year and though the idea is to protect the A10 crank maybe I'll get some feedback on taking it over 6,800 which it should easily do. An A70 crank is much stronger but probably very difficult to get.

The right hand mold is my Firebird port on the Thunderbolt head with 42mm valve, the best so far. At least of 34mm heads. The left one is a Lightning and it possibly 1 or 2cfm less.



This is side on. They are based on the XR750 port Jim shared but smaller. I think I can see what can improve the Lightning one a little.

@mark parker

very nice work, compliments!
Funny how results of work become quite convergent when one tries to squeeze out power for all its worth.
Your casts look very similar (from top pretty much identical, from side i seem to have done a deeper/steeper bowl LSR approach to the seat) to the one of the DBD34 that i did years ago.

Good day and all the best

Christian

Do these changes improve your performance at a very restricted rpm level or do they assist at all levels?
I ask because street riders are rarely WFO and look for more torque in the low to mid range.

Onder said:

Do these changes improve your performance at a very restricted rpm level or do they assist at all levels?
I ask because street riders are rarely WFO and look for more torque in the low to mid range.

Click to expand...

The answer is no, on the stock Firebird it will drive from 2000 because I was being very lazy yesterday in not changing to 1st it's best from around 3,000 but is after all only 650, when you turn the throttle on it responds instantly, if you used full throttle it may balk at that much at low rpm but in top just a bit of throttle around 3,000 feels very strong, and the more open the more strength, it feels very nice. I'm not sure the rpm full throttle is working best, it's hard to get it in first because rpm goes up fast as you roll it around, if you shift early by 5,500 or 6,000 and put 2nd on full or close to it it goes nuts, the best way I can describe it, and vibrates the bars etc as it goes past 7,000 but the rpm rise is slow enough to deal with, get 3rd to get into smoother territory but it's the same. The vibes really spoil it, because if they were not there at those rpm it has great zoom noticeable especially up hills and the vibes make you change The seat and pegs are ok it's mainly in the bars and the instruments blur. BSA balanced for low rpm not 6000, probably worrying if it was a bit lumpy down low people would complain. It doesn't want to give up pulling into higher rpm but I do because the fierce vibration feels like you should. But if it was smooth it should go to 8,500 - 9000 probably. As 90degree engines do that. Though this head is interesting because it adds power right through the midrange.

Most brit parallel twins seem to like 3k and above. My old 71 BSA Lightning was a real vibrator and it was stock. Always wondered if they final ones were just
thrown out the door good balance or bad. Except for the fact that the vibrations broke off large pieces regularly and killed lamp bulbs too it was actually reliable.
Rolling the throttle on or cracking it open a bit has always worked better than just WOT.

It strikes me as amazing that a bloke in a shed can achieve this when the BSA factory with all their knowledge from Gold stars couldn't. If BSA had done this they would have only needed one head casting instead of two. One thing is certain, if my A65 had performed anything like Marks I would still own it, but with the addition of a five speed 'box which I always felt it needed.
On the subject of vibration, mine was like riding a road drill until I had the crank balanced as part of the engine rebuild including the T.S. bearing conversion.

Onder said:

Most brit parallel twins seem to like 3k and above. My old 71 BSA Lightning was a real vibrator and it was stock. Always wondered if they final ones were just
thrown out the door good balance or bad. Except for the fact that the vibrations broke off large pieces regularly and killed lamp bulbs too it was actually reliable.
Rolling the throttle on or cracking it open a bit has always worked better than just WOT.

Click to expand...

BSA had their engineers study the vibration using around 10engines with different balance factors. They found 56% the best. It was smooth around 6,000rpm but it was a bit lumpy down low. So after the study and recommendation BSA naturally stayed at 70% on road/production engines and 56% on competition engines. They tried a 180degree and were suggested a 90degree but didn't build one, I have and they should have.

The result being my Firebird bottom end at 70% is dead smooth at 2000rpm pretty good at 3000 and starting to buzz at 4,000 and gets worse. So I have 21-47t sprockets so 70mph or a bit more is getting into vibes but you could cruise at that and maybe a bit more. In my opinion what it does at 2000 I don't really care if it was rough at 3,000 I wouldn't care if it smoothed out higher, usually the smooth area has smoothish areas either side. So I think I would prefer it smooth higher up in a more useful range because out on the highway at highway speed suits me and Being able to use 7000+ without the intense buzz would be much more useful.

The best solution is 90 degree. This is a 90degree 750. It's not dead smooth any more than a SV650, but possibly better than the old triple.



I have one 90 left plus a cam, it needs grinding and balancing, plus steel rods and ignition. It would be nice in this old thing, but it's nice to ride as is. And I'd like to try 56% and see what it does.

Darrrrrn, double post, friggin sausage fingers

First of my compliments go to mark for some really well executed work, being seemingly a Do'er.
@Matchless The guys at bsa, Imho knew that for sure but for reasons of production economy it most likely wasn't feasible to hand port every single cylinder head.
To my knowledge eg guys like axtell and regarding Goldie's perhaps even more kenny Augustine knew it as well.
But given that foundries back then were stuck with certain core techniques it was just not possible to adjust to everybody need.
Taking the topic a hint further, given the question regarding rpm range:
Of course it's better to adjust the porting to the riders or better engines needs.
Usually this is done either by looking at the flow area at the valve opening/timing area or in (i highly assume marks and my) the other case at looking at the restricting venturi diameter (that way in my very personal humble opinion) achieving higher velocities with less turbulent flow.
Important factor remains that the mean flow velocity should not exceed 0,5-0,6mach.
There are ways to calculate the precise, so to speak flow area by means of the discharge coefficient or taken even further by somewhat involved calculus procedures out of the flow time area's but with experience I'd reckon most do it based on seat of pants feeling or with approximation equations.

Kind regards and happy weekend to all

Christian

mark parker said:

BSA had their engineers study the vibration using around 10engines with different balance factors. They found 56% the best. It was smooth around 6,000rpm but it was a bit lumpy down low. So after the study and recommendation BSA naturally stayed at 70% on road/production engines and 56% on competition engines. They tried a 180degree and were suggested a 90degree but didn't build one, I have and they should

Click to expand...

Intresting the comments on engine balance factors for the BSA.
FYI R.H .Smith raced a Spitfire 650 ridden by Clive Wall and regularly won the Bantam production races championship, I've always wondered what they had as engine balance.
Also the British sidecar champion Chris Chris Vincent on what was reputedly a BSA 800cc!

Matchless said:

It strikes me as amazing that a bloke in a shed can achieve this when the BSA factory with all their knowledge from Gold stars couldn't. If BSA had done this they would have only needed one head casting instead of two. One thing is certain, if my A65 had performed anything like Marks I would still own it, but with the addition of a five speed 'box which I always felt it needed.
On the subject of vibration, mine was like riding a road drill until I had the crank balanced as part of the engine rebuild including the T.S. bearing conversion.

Click to expand...

This reminds me of the magazine road test that were printed in the 1960s, 70s they were always favorable. It strikes me they just rode a bike as it came off the production line and if it was a good one, it was handled over to the mag for that months road test, as for the rest that weren't up to scratch. . . . .

Matchless said:

It strikes me as amazing that a bloke in a shed can achieve this when the BSA factory with all their knowledge from Gold stars couldn't. If BSA had done this they would have only needed one head casting instead of two. One thing is certain, if my A65 had performed anything like Marks I would still own it, but with the addition of a five speed 'box which I always felt it needed.
On the subject of vibration, mine was like riding a road drill until I had the crank balanced as part of the engine rebuild including the T.S. bearing conversion.

Click to expand...

You were NOT buying a hand built race fettered motorcycle, so anybody who had a shed with the right knowledge could build a better, faster engine than was the norm.
For instance, anybody could just polish the cylinder head ports that would gain a bike length from a standing start