Reducing Squish on a Poulan Clamshell

[Chris-PA]

@Chris-PA
any progress on this?

I don't come here much any more so I didn't see your post - but funny you should ask! One of the inserts failed today, and I thought I should document that in case anyone else is tempted to use a 3D printed engine part.

I've put at least 8 tanks through it, probably more, and it has become my go-to saw for everything. With that GB bar and no spikes it has 19" of exposure, which is at least as long at a couple of my larger saws with 20" bars, and with the lo pro chain it is easily as fast.

Anyway, the engine isn't damaged and just needs a bit of cleaning. The insert that was still attached is interesting:

This end is on the exhaust side and is the leading edge where the counterweights approach the inserts. Notice how it is eroded and pushed back, though there is no sign of mechanical impact.

This is the trailing edge and the ledge is still mostly intact.


This tells me that the inserts were actually doing what I hoped they would - damming up the pressure wave pushed along by the front edge of the counterweights and creating a high pressure area right in front of the transfer opening. Still, it's pretty apparent that these materials are not appropriate for use in an engine. They are too low temperature and don't have the strength. I suspect the temperature is OK when the engine is running, but I pushed the heck out of this thing on 90+ degree days, and I'm assuming the heat soak was the problem. However, the erosion of the leading edge would have been happening while it was running. Today was not hot and I didn't run it that long.

I liked the way it ran too much to give up on this part of it, so I am considering a couple of options. One is to just skip the filler and make a close fitting aluminum edge on the front and back. I'm also thinking about how I could make fillers out of some material more durable. If I do that project it will probably be documented elsewhere.

 

[Chris-PA]

All better now.

 

[Moparmyway]

Now you went all out and blinged with JB ................... NICE !!!!!

 

[Chris-PA]

At the time I posted the pictures above I was not really participating here, so most of the info was put elsewhere.

I decided that the only appropriate material I have ready access to is JB Weld, but forming it was a problem. I wanted it to fit decently close to the crank all the way around the bottom cap. Then I thought that while the 3D printed parts might not hold up in the engine, they should work as a mold/form for the epoxy. I also decided to use the same cap I had modified for the first attempts, as it had pins inserted that would help hold the inserts. I cut some brass window screen strips to embed into the epoxy. I did one side at a time, after smearing the form and other parts of the cap with silicone grease as a mold release - which basically didn't work at all, so the form got destroyed each time. After googling it I found that car wax is better.

I had a lot of excess to clean/grind/chip off.

I cut the outer edges off an old cap so that I could clamp the bearings down and be able to measure the clearance to the crank - at least sort of.

Then I decided that the inserts really should extend further up toward the transfers in the cylinder, so I made another form to add an insert there. The car wax worked better on that one. I also did a little grinding to open the access from the barrier to the lower transfers. I did not take a picture of the mold form for the upper inserts, but this is what the initial cast looked like:

The ridges are artifacts of the 3D printing on the form. It turned out I made an error on the shape (I forgot to subtract the distance I moved the crank), so these had to be ground down a bit, and I never took a picture of that.

The intent of all of this was not really to reduce case volume, as the inserts are too small to matter much there. Rather it was to try to block some of the air pushed by the leading edge of the counterweights, creating a high pressure at the base of the transfers as they open. Basically, an inefficient and weak case blower helping to force air up the transfers. The thin con rod and narrow slot help some too. Of course I have no way to measure if it really does anything, but the saw runs well. It is holding a very consistent 10,500rpm in each of the cuts here:



I'm hoping it will hold together now, and I even bought a matching NOS Craftsman clutch cover so it won't look like such a jalopy any more.

 

[bikemike]

Very nice sounds responsive and strong. Is squish reduction something you would do again on other clamshell saws? I was about check up on this thread and ask what's going on ,then you popped In again.

 

[Chris-PA]

Very nice sounds responsive and strong. Is squish reduction something you would do again on other clamshell saws? I was about check up on this thread and ask what's going on ,then you popped In again.

I would do it on any of the Poulans with the rubber caps on the bearings, frankly because you don't need to be that accurate in grinding out the pockets. It was actually not that hard to grind the bearing pockets with a sanding drum on the Dremel if you go slow and carefully, and would be even easier if you just removed enough to tighten the squish. Factory is always around 0.040", so you would only need to take off 0.020" - I took off more here and made a pop-up, but that was probably not worth it.

On clamshells with machined pockets where steel bearings sit directly you'd need to use some sort of tooling on a lathe or mill. But these bearing pockets are used as-cast, and the rubber+sealant covers up a few sins!

Turning the cylinder wasn't much different from any other.

I learned that the later 46cc engines used on saws like the PP4620 should have the rubber covered bearings too, but my older 2775 does not.

 

[bikemike]

I would do it on any of the Poulans with the rubber caps on the bearings, frankly because you don't need to be that accurate in grinding out the pockets. It was actually not that hard to grind the bearing pockets with a sanding drum on the Dremel if you go slow and carefully, and would be even easier if you just removed enough to tighten the squish. Factory is always around 0.040", so you would only need to take off 0.020" - I took off more here and made a pop-up, but that was probably not worth it.

On clamshells with machined pockets where steel bearings sit directly you'd need to use some sort of tooling on a lathe or mill. But these bearing pockets are used as-cast, and the rubber+sealant covers up a few sins!

Turning the cylinder wasn't much different from any other.

I learned that the later 46cc engines used on saws like the PP4620 should have the rubber covered bearings too, but my older 2775 does not.

What was the stock comp. And compression after reduced squish. I'm thinking of trying it on my 4620 or a popup

 

[Chris-PA]

What was the stock comp. And compression after reduced squish. I'm thinking of trying it on my 4620 or a popup

I'm afraid I don't know - I don't own a gauge. I also don't think that low speed pressure readings mean much in relation to running cylinder pressures are rpm, which is part of why I don't bother.

 

[bikemike]

All that matters us that ur crapsman whoops azz and looks good doing it

 

[fordf150]

Strange that this popped back up. We were just discussing porting clamshell saws on another site.

Glad to hear it's still holding together and running good

 

[old guy]

I did one of my 2900s, these have the seperate bearing seals. Squish was .050 and compresion was 138
I decided to take .025 off the base and journals, I did it all with a dowal slightly smaller than the bearing wrapped with 80 grit sandpaper and a flat piece of formica countertop with 80grit on it, I have a lot of my depth mics and measuring tools left from my machinist days so with much measuring I I got the journals down .025 within .001 of each other.
After about 9 hours of sanding, turning & sanding and measuring, the base was down .025, this resulted in a squish of .025 and a comp. of 164.
I did a little porting while I was in there, (my first), I took .020 off the intake skirt of the piston and some off the flywheel key to advance timing.
I did not have the flywheel tight enough and when I tried to start it it kicked back and spun the flywheel off and sheared the key, I put it back on probably about stock position.
It started and I made a few cuts with it, it seems to have much more torque but it was running rich and it was too cold to mess with it.
Time will tell how it comes out, would I do it this way again? probly not.

John

 

[bikemike]

I like the wooden dowEl idea

 

[Chris-PA]

I did one of my 2900s, these have the seperate bearing seals. Squish was .050 and compresion was 138
I decided to take .025 off the base and journals, I did it all with a dowal slightly smaller than the bearing wrapped with 80 grit sandpaper and a flat piece of formica countertop with 80grit on it, I have a lot of my depth mics and measuring tools left from my machinist days so with much measuring I I got the journals down .025 within .001 of each other.
After about 9 hours of sanding, turning & sanding and measuring, the base was down .025, this resulted in a squish of .025 and a comp. of 164.
I did a little porting while I was in there, (my first), I took .020 off the intake skirt of the piston and some off the flywheel key to advance timing.
I did not have the flywheel tight enough and when I tried to start it it kicked back and spun the flywheel off and sheared the key, I put it back on probably about stock position.
It started and I made a few cuts with it, it seems to have much more torque but it was running rich and it was too cold to mess with it.
Time will tell how it comes out, would I do it this way again? probly not.

John

That sounds like a lot of work!

I considered cutting the bottom of the clamshell with a good flat file and just measuring on the corners, much as you did. I'm sure it would have been fine, but I had access to the lathe and was having fun 3D printing the mandrel.

You can get flywheels for the 46cc saws cheap on eBay if you want to replace it, but the key is not really needed anyway.

 

[old guy]

Yeah it was a lot of work, it took at least an hour to pull .005 off the base and without good measuring tools it is useless to even try unless you make templates like you did. I have another stock one just like it so it will be interesting to see how much good all this did, both will be running 14" bars & Stihl P S chain.
I got all my ideas from the threads you and others wrote on these forums, so thanks.

John

 

[bikemike]

Yeah it was a lot of work, it took at least an hour to pull .005 off the base and without good measuring tools it is useless to even try unless you make templates like you did. I have another stock one just like it so it will be interesting to see how much good all this did, both will be running 14" bars & Stihl P S chain.
I got all my ideas from the threads you and others wrote on these forums, so thanks.

John

Yeah you had some good running poulans at Wadena last year

 

[Chris-PA]

it is useless to even try unless you make templates like you did.

Surprisingly I found that the old bearing with the witness marks left in the rubber cover was a better depth gauge for grinding the pockets that the thing I printed.

 

[old guy]

I did it a little backwards, my depth mike has a square end on it so it will not give me an accurate read to the bottom of the journal, but I took a reading with it and added .025 to that and ground em down to that number, then I ground the base down till I had the original bearing depth again. By the time I got done the names Bridgeport & Southbend were were looming big.

John

 

[Jeff Lary]

Another video - I wanted to try it out with a more typical bar length rather than the oversized GB bar. This is a bar from a McCulloch that is supposed to be 18", but is actually 17". I got it on a used saw and had to modify it so it would oil, otherwise it was similar to an A041 mount bar. It's 60DL rather than the 62DL typical on Poulans. The chain is WodlandPro (Carlton) lo pro.



I'm really happy with the way this thing runs. I put the GB bar back on it after a while and have decided that it pulls the longer bar just fine, and oils it well, so I'm leaving that on. I have other lighter saws to use the shorter bars on.

I am coming up from behind here but Chris you did a real good job the saw cuts that Oak like white Cedar.

Edit:
Well I just finished up , I read all of this thread as my nephew has a couple of the wild things I think and you have inspired me if I can get my hands on them . They will not run god only knows why ,..but any who....
Chris in the beginning unless I missed it you did not say how you measured the squish to start with? Did you use solider? if so was it plumbing solider? my electrical solider would not be thick enough to measure a big void like you had. Also I don't have any lathe,.. but I wonder if I could " draw file" the cylinder base down some or sand it down using some sand paper stuck down to a block of Granite countertop I have? Well tanks for your advice in the past and future, you did so cool stuff there ! Jeff

 

[Chris-PA]

I am coming up from behind here but Chris you did a real good job the saw cuts that Oak like white Cedar.

Edit:
Well I just finished up , I read all of this thread as my nephew has a couple of the wild things I think and you have inspired me if I can get my hands on them . They will not run god only knows why ,..but any who....
Chris in the beginning unless I missed it you did not say how you measured the squish to start with? Did you use solider? if so was it plumbing solider? my electrical solider would not be thick enough to measure a big void like you had. Also I don't have any lathe,.. but I wonder if I could " draw file" the cylinder base down some or sand it down using some sand paper stuck down to a block of Granite countertop I have? Well tanks for your advice in the past and future, you did so cool stuff there ! Jeff

I don't think I said how I measured it, but I always use solder - usually 0.032" electrical solder which is pretty soft. Typically I will twist two strands of that together so I can measure squish at 0.040" like this one. I've got lots of 0.064" solder around too but never tried it.

I think you could do this with a sanding drum on a Dremel for for the bearing pockets (note there is a ridge on the outside to position the bearings/crank), using the old bearings as a gauge, and a good flat file and/or flat table and sand paper for the cylinder. If you look at the picture below I measured the thickness at each corner so I could verify how even I got it. Also, the factory cylinder is not machined on the bottom, so it will have a fair amount of variation stock.


This shows the bearings in the cap, but they fit the same in the cylinder and the old center seam is clearly visible on the rubber:

 

[LegDeLimber]

I'm afraid I don't know - I don't own a gauge. I also don't think that low speed pressure readings mean much in relation to running cylinder pressures are rpm, which is part of why I don't bother.

Looking at those "rubber" bedded bearings and wondering how much of a dynamic squish band is in play here also.

First thought of this as a bit of flippant self mussing and then said to self, hmm tis likely far truer than you'd expect.