valve timing help needed!!!

Hi, Im building a 3/4" Holden traction engine and I cant seem to get the timing right? Ive had different suggestions but none seem to work or maybe Im thick??

One was to get the eccentrics 90 degrees to the crank pin,that I done but it jams up when turned over before the eccentrics have gone full stroke?

The ins. with the kit are a joke but the model is well made.

Thanks in advance.

Terry

[This posting has been removed]

Set eccentric's high point at 90deg + 30deg in advance of the crank in the direction you want it to turn which is a good starting point

The eccentric throw reads a shade too much , that should never happen.

As Jason says, but the OP seems to be not too far away at 90 deg. In that case I'm surprised that the gear jams up, which suggests something might be wrongly dimensioned.

How do I adjust the eccentrics when the grub screw is buried under the strap connector/bearing?

[This posting has been removed]

There at the top.The grubscrew get buried when turning the eccentric.

[This posting has been removed]

Well if hole is at the top and grub screw usually drilled through the high point of the eccentric poke an allen key into the hole and loosen the screw and keep key pointing upwards and then set crank to the angle I suggested which will either have the throw at 4 0'clock or 8 o'clock depending on direction of rotation of the crank for that particular eccentric. Then tighten screw and give it a try.

Can you post some photos of the eccentric crank and valve / piston end in various positions Terry? Perhaps showing what it looks like at each stage of Jason's instructions and where the jamb is. The problem might be obvious in a photo: words are less reliable!

Dave

[This posting has been removed]

Dave, I would imagine that the "jam up" is more likely the piston simply being pushed straight to one end of the stroke by poor timing rather than a physical tight spot in the valve gear.

Don't forget that if it is a 4 shaft engine the flywheel turns "backwards" when in the forward position on the reverser. eg anticlockwise when viewed from the flywheel side.

Edited By JasonB on 28/01/2021 17:00:08

Check out Keith Appleton on youtube, he has several videos on valve timing.

Right, assuming reasonably conventional engine with Stephenson's Link Motion...

'

The jamming does not sound like a timing error but might be:-

- The piston would seem to be right, but the valve may be a little out of position so hits the steam-chest wall. You can test the piston by rotating the engine with the eccentric-straps removed and well clear. Usually, the valve-spindle is adjustable and we may need visit that later.

- The eccentrics being in some odd position that makes the expansion-link edge meets the back of one or both eccentric-rod clevises or that of the valve-spindle.

- The die-block is hard against the end of the slot in the link. I am not sure what effect that would have, but it won't help.

Faults 2 and 3 might arise from the eccentrics not being where they ought be. Maybe that 90º? It would be a very unusual engine designed for that.

'

First, ensure the access-hole for the grub-screw is accessible then set them [90º + about 30º] or a little less, as Jason suggests, but ahead of the crank in their appropriate directions. There are techniques for doing this accurately but for the moment making their pattern sensibly symmetrical about the crank is the main thing. That extra angle, is the 'angle of advance' and subject to further adjustment.

Remove the valve-chest cover if already fitted.

Put the engine in mid-gear (minimum valve-travel). Gently rotate it to ascertain what if anything is still jamming where. Also see if the ports open. They should not, but if one does, I'd first consider the valve-spindle adjustment.

Repeat the test, this time in both full-gear positions. The ports should open by the same width, which might not be the full port width, depending on the design. It's symmetry and of course, free running without binding, that matters here. You can measure the openings by feeler-gauges or similar.

I don't think the port-openings will be necessarily equal in reverse to forwards, as that could be affected by the way the expansion-link is suspended. Traction-engines reverse only for turning and parking - get it as close as possible but the forward gear timing is the more important.

Now - the subtle bit. That angle of advance draws the valve back to start opening the port as the crank runs over dead-centre. Normally, a 90º only setting means the valve would still be covering the port by the lap width.

Set full forward gear, turn the engine in the appropriate direction and stop it on dead-centre as close as you can - as I say there are established methods for this, but depend rather on the engine's design.

Assuming no lead, the port should be just on the point of opening. (Lead is a tiny opening only just before dead-centre, but your engine may not have that, by design.). If it isn't, adjust the eccentric that's in line or nearly so with the valve spindle, until it is.

Turn the engine slowly to max. port opening: measure and record it.

Continue to the other dead-centre, noting on the way that the first port closes at designed cut-off, (maybe around 3/4 stroke); and see if the second port is just about to open as above. Carry on round until you've completed the circle and the valve-travel and port-openings are as close to symmetrical as possible. Adjust that direction's eccentric as necessary. An error here may be from the other eccentric being still out of angle.

Repeat this time with full back-gear.

Tweak both eccentrics and if necessary the valve-spindle again, until the machine is as close as possible.

Essentially the whole thing should be all symmetrical, but you may find some difference between ahead and astern running if the expansion-link's suspension is at one of its ends. (It usually is on traction-engines. On a loco it's in the middle for equal running both ways.).

'

If things still jam and we are sure it's not the piston, crosshead or valve hitting something, I would suspect a faulty dimension somewhere; probably around the expansion-link area.