Messages

Anybody want to chime in on the pros/cons of decking 350 cylinders so you can use a 400 gasket? Some very reputable people (some on this forum) have suggested this as a way to go, and with no disrespect to them, just want to gather more info so I can decide if this is a mod I want to do and if the benefits are worth it and if there are any significant cons. I'm still going for reliability over pure power so curious what people think.

I've done this on a couple - it's a 5 minute job with a fly cutter on the mill. I like it as you can do it in conjunction with setting the squish. Worth noting that if you want to set the squish tighter than about 0.9mm, you want to pin the heads to the barrels so it stays consistent. I've done this with two bits of drill rod through both heads into the cooling fins on the barrels which is not the most accurate way but works very well. You take them out once you torque the heads on.

I'd buy 250 heads as they are way cheaper and pay someone to recut them - it's not a hard job and you get all the benefit of a real performance head for a fraction of the price. I've done them at 50/45/40% squish area and various CCs. I think 21cc would work on a stock(ish) bike, but I had to go larger on a ported engine and good pipes as it was detonating. Think I ended up at 23cc. Squish somewhere between 0.7mm and 0.9mm has always worked well. Stock is insanely large, like 2mm, so there's loads of room for improvement.

I've had a set of Kveldwulf cylinders up and running for a bit - they work great once cleaned up. Mine came honed quite large for some of the cast pistons but worked perfectly out the box for pro-x. Definitely measure first. I've never run the stock pistons they ship with.

If the main bearings feel silky smooth and are still covered in oil, I'd just change the oil seals (crankshaft and gearbox output shaft) and bung it back together with the yamabond. Give all the dogs in the gearbox a quick once over too.  You could measure the run out on the crank if you feel like. I have had good luck with the new Taiwanese aircooled cranks you can find on ebay - the last one I bought was about 400bucks and insanely true. That said, it's still in the box as the current 250 crank that's in the engine keeps humming along smoothly.

PWK jet sizes don't correlate to Mikuni. 210 is a whopper for Keihin.

2T5 - that looks OK to me mate, if that was mine, I'd probably try one size leaner. Going richer first can't hurt though.

Is that from a single pass? If so, the pic on the left looks plenty rich to me. I've always aimed for a consistent, milk-chocolate brown line a couple of mill up the insulator.

When I'm not sure on jetting (which is regularly :)), I go up a BUNCH of sizes until it chokes up noticeably on the pipe, then come down one at a time until I'm happy.

on board starter and lights   

Haha - amazing.    What a blast that would be in the canyons...albeit a bit thirsty in pistons!

Nice! A 2 stroke race bike that idles, whatever next? :)

Nice, I'll keep following this one. Hopefully inspiration to sort mine...

Cheers.

Nice, excellent to see it making noise again! Are you going to give it a strip down for seals, etc?

Totally agree, with m in sc. Without powervalves, best you'll get is about 3k rpm of usable power and some overrev. Where you want it, depends on the ports and pipes. It's difficult to shift the hit much below 5.5k though, because the stock port timing is already pretty racy.

 FWIW, I suspect that dyno is not a like for like comparison under 4k, good chance he's only opened the taps at 4.5k or so. Mine is nice and tractable from about 4k, but it's not going anywhere fast until just under 6.

I grew up roadracing 125s so 6-9k IS midrange for me!

That R5 sounds sweet!

Have a peek on Mick Abbey's Facebook page. He's a UK tuner that does a lot of vintage stuff and almost always posts dynos of his results. He's an absolute gentleman and has given me a lot of really useful tips. This is for a 350, and was what I was aiming for with mine...

https://www.facebook.com/photo/?fbid=2703540803238979&set=pcb.2703540876572305

engmod is a very sophisticated engine modelling software - you can model everything from intake through to exhaust as a system.

I'd think about the whole package before you get cutting. 350 barrels need the lip skimmed from the top, for o-ringed OEM heads, or RD400 gaskets and you kinda need to do that before you can set the squish gap. Also, total head cc is going to be dependent on the rest of the tune (porting and pipes.) A stock 350 has a head volume of about 22cc and runs fine on pump gas. Add good porting and pipes and it probably won't, as the dynamic compression goes up from the effects of the tuned pipe. Mine had detonation @ 22cc / 0.95mm squish gap. Still at 23cc / 0.95 and eventually cleared up at 23.5cc and 0.85mm. An experienced tuner will have done all the experimenting already.

I like the HPI ignition which is relatively cheap but seems to work well. Doesn't have a lot of output for lights but works fine with LED bulbs.

That paint is awesome - nice job!

Those billet heads are really nice. I'm going to find a pair of late model Euro 250 1A0 heads next time - they have an extra fin and are for long plugs, so you have a lot more to play with. In a, probably rash, move - I decided to bite the bullet and ordered the cheapest, taiwan made, 4mm banshee crank I could find yesterday. Will let you know how I get on with squeezing it in...

There's been a lot of research done on squish velocity, apparently the numbers to shoot for are something like 20-29m/s at peak power RPM. If peak power RPM is lower, you need a greater squish area to achieve it (squish angle also is a factor.)
Fortunately for us plebs, clever people have been experimenting with RD350s for decades, so anything 45-50% is going to work well. I'm pretty sure I set my compound angle to 14 degrees (will check notes later), which should be fine for RD/YPVS pistons.

250 heads are good because they're way cheaper and more plentiful than 350 heads and you have a little more metal to play with. The only prob with all US 250 and 350 heads is that they're all for the short plugs, so to get the sort of chamber volume you need for pump gas, you might have to put spacers under the plugs to pull them up a tiny bit once you're done.

This is how I do them. Pick out a pair of vaguely matching heads and lap the faces on a surface plate, just to clean them up. Then measure from the top of the plug. Put them in the mill and clean up the top of the plughole and head bolts until they are equal.

Make an arbor to hold the heads from the plug hole, which is just a piece of mild steel, machined down to 1" diameter with a 14 x 1.25mm thread on the end and a centre reference. You can use 5c collets on my lathe, which hold everything  accurately and are handy for this.

Once it is all mounted up, spin them up and clean up the gasket land, about 15 thou on each. After that, you can measure the height from the centre reference and work out the hyp, to scribe an accurate OD on the head for where to start cutting the squish.

After that, set the squish angle on the compound and cut until you hit the OD. It doesn't matter how wide it is at this point as you need to recut the bowl anyway.

For the next bit a radius turning tool is helpful. I worked out that a 28mm radius would be a good starting point for a bowl that would intersect the squish band about the right point (you can mark up the id of the squish on the part with a pen, same as before. This gives a single radius bowl of a little over 12cc total vol, which added to the vol under squish (about 5.5cc) and the tiny bit in the plug recess makes for a hair under 18cc. This might might work on 100LL, which could be fruit for a future experiment as it's a doddle to cut.

For pump gas you need more vol (how much exactly is a bit of an open question, depending on your gas / altitude / tuning etc) but I'd say min 22cc, which is about what the stock 350 heads have. I cut this a bit at a time with a large radius tool bit, using dyechem to mark up the bowl, a bit like wood turning. You need to stop and measure with a burette regularly, but you get a feel for the cuts and the vol and it doesn't take long.

Here's a couple of pics of the tools / process.

I got the bog stock HPI system which only has one curve - you can check it out here:

https://www.hpi.be/item.php?item=210K176

Seems to work alright but people do make decent gains dialing in a custom curve. I love the little flywheel.

I want to try making a 4mm stroker next time I build one. You pick up a load of time area from the longer stroke, which is tough to do with the limitations of the castings at 54mm. On a 65mm Bore, you'd have an RD385 and probably a genuine 60+hp somewhere between 8-9k, without crazy porting. Pretty sure it's possible with some fiddling around....

I'm just a hobbiest, but this is what I did, I'm sure others might do it similarly - you could probably go 1mm higher and not as wide on the exhaust and get similar and perhaps better ring life. I've not tried back to back.

I raised the exh port to 28mm and widened it to 43.5/44. Raised the transfers to a 42mm and milled out the roof of the little one as much as I dared to open it up and match the roof angle of the A transfer better. I also redirected them as best as I could to get the streams closer to Jan Thiels model. Cut back the base of the main transfer by about 8mm, and also got rid of the hump in the tunnel. Raised and opened up the boost port to 18mm and pointed it just in front of the spark plug. I skimmed the barrels for 400 gaskets and pinned the heads on centre. I recut 250 heads to give 45% squish by area and 0.85mm clearance - total head volume is about 24cc.

Running aftermarket chinese cylinders, prox banshee pistons, standard reed blocks with YZ85 reeds, banshee inlet rubbers with link pipe and stock 28mm carbs with the Dale F mod. Single large foam filter on the standard intake rubber. HPI ignition.

Made pipes to a design by Mick Abbey in the UK - It was quite close to what I came up with on the 2t exhaust calc sim, but with a few notable differences. The overall length was almost identical. You want pipes for an 8.5k peak or thereabouts. Speak to Mick or someone local here in the US and tell them what you want the pipes to do.

I've not dynoed mine but it's smooth all the way through to 9k and will pick up the front, smartly, in second from about 6k. It's a really nice engine.