The Lift Plate Engine Build
The Build
Very nearly two years ago I started a project to rebuild the engine of my 1976A test mule by doing some dyno testing. This was to establish a baseline of how the engine was performing before actually making any changes - I wrote about it here. The work kept being put to one side as life got in the way, but recently the job was done and this morning I re-ran the dyno testing on the fresh engine and we have a result ! First though, a bit of a refresh.
This bike is a basically a test rig, and over the years I have hung various bits and pieces on it - it was for a while a very convenient place to store good spares for my other 1976A, it has had a couple of different engines installed for testing, as well as several carburettor banks, again for testing. This time I wanted to try something different, and after some discussion with a friend of mine in the USA decided to do a two-step tuning process:
Step one would involve seeing what difference a lift plate by itself made to the engine performance
I wanted to then see what difference various exhaust systems actually made on the new fresh engine
Important note: I am not any sort of tuning expert. There are many people out there who will gladly step forward and make that claim - I do not. Secondly, the few people I do consider to be 'expert' also are trying to make a living preparing competition machines, and so do not generally share exactly what it is they do to an engine to make it perform ! As a result getting any genuinely useful information is not easy. For this reason, whatever level of performance I have been able to achieve I know can probably be much improved - all it really takes it money.
The Suzuki GT750 engine can be made to produce in excess of 100 HP, but that level is not really suitable or safe for a road bike of this vintage still using a stock frame and stock brakes. For usability, a target of 70 to 80 HP is commonly seen. I am not going to go into the theory behind the 'how' part - there are lots of books out there that describe this, but to get that sort of power with a 2 stroke engine normally means having to do two things: change the porting to make the engine breath better, and then use a tuned exhaust pipe to hold the fuel charge in the cylinder long enough for it to generate more power when combusted.
Back to Top
Lift Plate
One of the problems with the later GT750 engine (post 1974) is that to open up the exhaust ports risks cutting through into the cooling water jackets on the cylinder blocks. For this reason a 'lift plate' is commonly inserted under the block to raise it and the ports up, and then the amount of rise is milled off the top of the block. I used a 2.5mm lift plate supplied by Cometic in the USA - it is essentially an aluminium plate cut to match the top mating surface of the crankcase, and is seen in the photo below to the left (just click on the photo to see a larger version). After adding in the gasket thickness I had a bit over 3mm milled off the top of the block, such that the piston outside edge was even with the top of the block when at TDC - the technical term for this is to have a zero deck height. The cylinder head was also skimmed - and this caused a bit of a problem as when I fitted the head to the block it rocked slightly. To correct this I lapped the head to the block and checked for high spots using machinists blue dye - shown in the centre photo. The lift plate fitted in place is shown in the two right side photos.
Back to Top
Porting
Note: I did not do any actual changes to the size or shape of the ports themselves, as at this point what I wanted was to determine what if any change was attributable to just using the lift plate.
Using the lift plate effectively raises all the ports by 3 mm and the expected net effect of doing this is a general increase across the power band. Specifically I'm told that raising the exhaust and transfer ports should improve mid-range and top end power in the 5500 to 8000 RPM range, and raising the intake port will help power below 5000 RPM. And that is without doing anything to the shape and size of the ports themselves.
I did clean up the castings themselves and polish the exhaust and transfer ports. I also matched the barrels to the upper crankcase as there were 'steps' in the fitted assembly. This can possibly be seen in the photo below left. In the centre photo, you can possibly see the casting steps, as well as the SRIS injection nozzle. I smoothed out any steps cleaned up the rough edges and removed and plugged off the SRIS ports.
The inlet spigots that the rubber boots mount to were especially mis-matched, and a lot of material was removed. The before photo above right shows the size of the step. As well there is the divider on which there are various views. The later M/A/B model blocks have long dividers, and the earlier J/K/L blocks have short ones. Richard Nowsen contributed a good description with photos which I have included in the GT750 FIeld Guide at this link. Later GT750s supposedly developed more horsepower than earlier ones, but had the longer dividers, however many 'experts' will tell you make the divider as short as possible. And that is without getting into the debate about making it knife edged, or slightly rounded, highly polished or slightly rough. I split the difference on this iteration and shortened the divider, but not as far as is seen on the J/K/L blocks. I also did not knife edge the divider, nor did I highly polish the inlet tract. I left the oil injector nozzles intact.
Back to Top
Closing Up
A few final items to mention:
- The crankshaft was rebuilt using Suzuki parts (seals, pins, bearings, rods and first over pistons). As seen in the photo to the left, I use old crankcase halves to store crankshafts - it just ensures the bearings are properly supported.
- I installed an aluminium water pump gear - mainly as I happened to have it, and it looks nice - see the second photo.
- I replaced the points ignition with an Accent electronic unit.
- I reused the stock BS40 carburettors, and left them on stock jetting. This means the engine runs a bit rich at our altitude (Calgary is at about 1084 metres, or 3557 feet above sea level)
- The compression is 150 PSI across all three cylinders - I'm running NGK B8EV spark plugs, stock 45 year old coils and stock timing.
- As I wanted to compare against stock pipes, I had mine chemically cleaned, and used an inspection camera to check and see just how clean they were (not very, but better than how they looked before cleaning). A photo is on the right.
Back to Top
The Result
So the day had come - after all the hard work not to mention a chunk of loose change, had these engine modifications actually improved anything ? Well, for one thing you will note there was a lot less smoke than the first time when the benchmark testing was done !
After running in the engine and putting a few hundred miles on it, I took it back to RPM Services south of Calgary for another dyno day. As well, I took along a second exhaust system made by Gibson Exhausts in the UK. I bought these a few years ago when living in Europe as I just liked the look of them, but I also always wondered whether they actually did anything to improve the performance other then weighing a fraction of the stock exhaust system.
After several runs, the results were in - the chart above shows three curves - the lowest one is the result from the bench mark run before the engine rebuild. The middle line shows the result from the rebuilt engine fitted with a stock exhaust system and also with stock BS40 carburettors and airbox. The carburettor jetting is also factory stock, which for our altitude here in Calgary means the engine runs a bit rich which is OK as a starting point. The top curve shows the rebuilt engine fitted with the Gibson exhausts swapped in for the stock exhaust system, but with no other changes (so stock jetting on the BS40 carburettors, etc.).
You can click on the photo above to see a larger version, but to recap - the numbers are 55 HP at the wheel, dropping off to 53 HP in the original engine as it approached 7000 RPM. The rebuilt engine with the lift plate and clean-up (but no porting) and stock exhaust, BS40 carburettors with stock jetting, airbox etc., managed a clean performance increase right across the rev range topping out at just under 63 HP. That is an increase of roughly 15% which is respectable. The top curve shows that while slightly less than the stock exhaust below 4000 RPM, the Gibson's come on pipe at about 4000 RPM and then peak at 70.2 HP at 7000 RPM, for a performance increase of roughly 27% over the base case, which is quite respectable.
So what's next ? I have a couple of other exhaust systems I plan to test with this engine in its current iteration, and will eventually post the results here. At some point in the next year or two I'll pull the engine and conservatively modify the porting, just widening the exhaust and intake ports, possibly dropping the exhaust port floor slightly but not much else. Then I will do the dyno runs again with the same sets of exhausts and see what changes.
And why bother doing any of this ? No reason really - having long winters allowing lots of shop time I suppose is a part of it, but truthfully ? It just makes me smile !
Back to Top