GS450 Low Compression after rebuild

[Guest]

I've been rebuilding a GS450 for the past 2 months. I finished (or so I thought) rebuilding the top end of the engine, and now I have bad compression. I don't know what it was like before, I never saw the bike run. I swapped the valves from my old head into the new, and I'm wondering if that could be my problem. I replaced the cylinder gasket, head gasket, stator cover gasket, cam cover gasket, breather gasket, replaced valve seals and shims, and cleaned and painted the cylinders and head. I put it all back together and had a leaky carb, so I drilled and tapped for bigger screws on the bowls and that solved that problem, but the bike still wouldn't start. I just rented a compression tester and found that the left cylinder managed to put out 70PSI, while the right put out only 40. How can I easily go about troubleshooting this, and could timing be an issue? I had to move the crank forward and then back just the slightest bit to get the chain unbound when I reassembled, and now I'm wondering if that threw off my timing.

[gs45082]

Are your valves adjusted to 0.03-0.08 spec? I had terrible compression on my 1000 caused by tight valves.

Did you lap the valves in when you installed them?

[JRHemmen]

Yea, they're all within spec now. I didn't lap them. I hope I don't have to take them back out, I'd give anything to not have to deal with advance auto's valve compressor again. Swapping valves took me over 6 hours.

[tkent02]

What method did you use to break in the rings?

[JRHemmen]

What method did you use to break in the rings?

the rings are the one part I didn't replace in the rebuild. Should I have?

[Nessism]

You don't sound very confident about the cam timing so I'd start there.

You should have lapped the valves at minimum, and possibly replaced the rings and honed the cylinder. The second part depends on whether you knew for a fact that the engine had good compression before the teardown. That's the only way I wouldn't change them.

[JRHemmen]

You don't sound very confident about the cam timing so I'd start there.

You should have lapped the valves at minimum, and possibly replaced the rings and honed the cylinder. The second part depends on whether you knew for a fact that the engine had good compression before the teardown. That's the only way I wouldn't change them.

I just watched a video on lapping valves, I'll try to get the head pulled and do that sometime this week. The PO said he had it running last year, but that he couldn't get the air/fuel mixture right. He put on pods and chopped the mufflers without changing the jets, so I'm assuming that his problem was purely in the carb and that his compression was okay.

Will lapping have some affect on the shim clearances?

[tkent02]

Before you take it apart, ride it a while and see if the compression comes up. Hard to say with old rings in an old bore, but it might help.

And, you didn't mention it, did you hold the throttle open while you did the compression test?

[JRHemmen]

Before you take it apart, ride it a while and see if the compression comes up. Hard to say with old rings in an old bore, but it might help.

And, you didn't mention it, did you hold the throttle open while you did the compression test?

I don't think I'll be doing too much riding without having started it first, hahah. And no, I didn't touch the throttle. I have the carbs and exhaust off now as I had some other work to do to the engine. Next step is to replace the stripped studs on the oil filter cover, then lap the valves. How hard is it to get the head out while still in the frame?

[jdvorchak]

You should have lapped the valves anytime you remove and replace them.

We can guess at this all day. When I'm stumped I drop back and pretend I've never seen the bike before and begin trouble shooting from square one. The only way to diagnose the problem is with a leak down tester. I built one for under $20 with a Harbor Freight air regulator some air fittings and a pressure gauge. You may be able to get a loaner from O'Reilly, Autozone etc. Do some searching on youtube for instructions on how to build one and how to use it. But first make sure your cam timing is correct.

[JRHemmen]

I took the cam cover and cams off this morning before class. I'll be lapping the valves and redoing my timing, two birds with one stone I guess. Hopefully after this I'll finally be able to start the bike. I wanted to at least see it run before I poured more money into lights, bars, and everything else.

Do I need the specific valve lapping compound, or will any regular lapping compound do?

[gs45082]

I took the cam cover and cams off this morning before class. I'll be lapping the valves and redoing my timing, two birds with one stone I guess. Hopefully after this I'll finally be able to start the bike. I wanted to at least see it run before I poured more money into lights, bars, and everything else.

Do I need the specific valve lapping compound, or will any regular lapping compound do?

I used "valve lapping compound" from autozone I think. Lap them in until the seating surfaces are smooth and the correct width (I think a mm or so, ninja manual has a spec). When they're all done, wipe every last bit of compound off the head and valves. Check your work by installing valve springs and a spark plug, lay the head upside down, and fill the combustion chamber with oil and letting it sit on your bench overnight. If one of the ports is wet from oil re-lap that valve.

[tkent02]

I don't think I'll be doing too much riding without having started it first, hahah.

Why not put it together, check the camshaft timing, start it up and ride it? Something I'm not understanding here.

[JRHemmen]

Why not put it together, check the camshaft timing, start it up and ride it? Something I'm not understanding here.

The bike was unable to start in its current condition. I'm sure the valves are leaking, but I think the timing should have been fine. Now I'm taking it apart to lap the valves and make sure the timing is right. Hopefully after this I'll be able to actually start it.

[tkent02]

That compression problem by itself isn't enough to make it not run. Something else is wrong.