Riders Of Vision

General => TechTalk => Carburetors => Topic started by: pullshocks on March 01, 2014, 05:48:06 PM

Title: Solution for float level adjustment vexation (?)
Post by: pullshocks on March 01, 2014, 05:48:06 PM
Picking up where I left off last fall, http://ridersofvision.net/rovforum/index.php?topic=14994.msg138170#msg138170 the new petcock is working fine, but still getting  overflow from the rear carb.  I replaced the needle valve o ring, and the needle and seat seal nicely, but I am having trouble getting the float adjusted right...or the float is defective, or.....if only I could see what's going on inside the carb......hey, what if I....
Title: Re: Solution for float level adjustment vexation (?)
Post by: The Prophet of Doom on March 01, 2014, 08:16:09 PM
I had a LOT of trouble with the float levels on the 400.  Even with a working petcock there's enough fuel just in the hoses to make a decent sized puddle on your garage floor.  I would set the level correctly, then assemble it would seem OK but then overflow once on the bike - especially on the side stand.

Lucky observed carbs with the overflows a couple of mm lower than they ought to be.  It's worth measuring yours.  Mine were about 15mm below the gasket from memory, but Lucky suggests 13mm as correct - I guess that made mine more susceptible to an overflow.

There are two float settings - one that measures the float height 36mm (you measure this with a ruler Haynes 12.9a), and the actual fuel level 20mm (You test this by putting clear hose on the overflows and opening the drain screw Haynes 12.20)  What I found was that although the float height was set correctly, the floats weren't as buoyant as they needed to be to end up with a  fuel level of 20mm.  Although the fuel level was below the overflow on my bench, heaving the bike around or putting it on the side stand caused an overflow. 

Testing various floats I did find variation in floatiness so the ruler measurement is more of a starting point than a setting.   The actual setting should be entirely on fuel level - it took a few goes.  And since the spec is 20mm +/- 1mm I set mine to 21mm to best handle my low overflow pipes.
Title: Re: Solution for float level adjustment vexation (?)
Post by: pullshocks on March 12, 2014, 11:16:07 PM
OK the transparent box is part of a test system, set up shown here.  Fuel resevoir at top, catch basin at bottom.  This is an 83 carb with the fuel return line.  The "return" fuel is allowed to simply accumulate in the catch basin.  Results in next post
Title: Re: Solution for float level adjustment vexation (?)
Post by: pullshocks on March 12, 2014, 11:38:08 PM
OK.  Filled the reservoir with fuel, turned on my little plastic NAPA in line fuel valve.  Because the opening in the needle valve is quite small compared to the diameter of the return line, most fuel runs down the return line, and the transparent box fills up very slowly.  The float eventually does its job, and  the fuel levlel stabilizes.  It doesn't show up too well but I have scribed lines at 19, 20, and 21 mm.  On my first try, the level is well below the 21 mm line.  Too low, but at least I can verify the needle valve shuts off.

More int he next post.
Title: Re: Solution for float level adjustment vexation (?)
Post by: pullshocks on March 13, 2014, 12:26:03 AM
Of course, the "official" float test is done with the engine running at idle, and the above is just  a static test.  To simulate fuel use at idle, I included a valve in the bottom of the transparent box. 

I reckon that fuel consumption at idle is a fraction of an ounce per minute.   If you get 50 MPG at 50 MPH, that is a a gallon per hour, or about an 2 ounces per minute divided between 2 carbs, or about an ounce a minute per carb.  At idle it would be much less.  So I opened the valve enough for a drip every few seconds.

I found that the float and needle valve did maintain a steady fuel level.

One flaw in this test is that fuel pressure and flow from the fuel pump may be greater than the pressure due to gravity in my test system.  A higher pressure would require a different float setting to achieve the same fuel level.

I need to investigate this by capping off the return tube so all the pressure bears on the needle valve instead of relieving itself via the return line.

One more thing--POD is right-- the ruler setting is just a starting point.  Especially if using the needles in the Keyster rebuild kit, which are shorter than the originals (at least in the '83 kit, don't know if this is true in the '82 kit)
Title: Re: Solution for float level adjustment vexation (?)
Post by: Rikugun on March 13, 2014, 11:41:46 AM
Interesting experiment! Has it showed an overflow condition yet? You can also raise fuel pressure by increasing the distance between the fuel level in your reservoir and the inlet valve.

That's pretty nice solid core plywood for a test fixture.  :)
Title: Re: Solution for float level adjustment vexation (?)
Post by: fret not on March 14, 2014, 02:02:01 AM
Yeah, I was admiring the fixturing also.  Nice work.

What did you use to seal the Lexan box?
Title: Re: Solution for float level adjustment vexation (?)
Post by: pullshocks on March 15, 2014, 02:11:23 AM
Strictly  bits of scrap plywood from the  pile.  My shop will be featured on the new reality TV show"Woodworkers who hoard."

Enclosure was sealed with Seal-All, available at auto parts stores.  Very effective but messy to work with.  Of course with my matchless fabrication skills, a sealer was hardly needed.

Thanks for the tip about raising the supply can to increase pressure.  I am very curious to see how much of an effect this will have.

Title: Re: Solution for float level adjustment vexation (?)
Post by: pullshocks on March 15, 2014, 05:46:50 PM
The best I could do was to raise the supply tank about 6 ft above the needle valve and cap off the return line.  At that pressure the fuel level stayed pretty much within the desired +/-1 mm range.  One time it went up to 18 mm below the gasket, but I never got it to do that again, so that might have been a bit of crud getting in the needle valve.  In real life I use a fuel filter, of course.

The overflow tube in this carb is a whopping 16.5 mm below the gasket, so going over bumps and what not, I can see it would be easy to get a bit of overflow.   The float pivots up and down in a fore&aft direction, so up and down hills may affect fuel level somewhat.  Leaning side to side theoretically should not affect it much,  I don't know why P.O.D. would necessarily be getting overflow with the bike on the side stand, unless the float or needle valve was sticking for some reason.

I am going to take the same approach as P.O.D. and set the float level at 21 mm in this static test, hoping to keep overflow episodes to a minimum under use conditions.
Title: Re: Solution for float level adjustment vexation (?)
Post by: QBS on March 15, 2014, 06:34:05 PM
If all else fails, you can always install inline fuel cut off valves.
Title: Re: Solution for float level adjustment vexation (?)
Post by: The Prophet of Doom on March 15, 2014, 07:05:07 PM
Quote from: pullshocks on March 15, 2014, 05:46:50 PM
I don't know why P.O.D. would necessarily be getting overflow with the bike on the side stand, unless the float or needle valve was sticking for some reason.
Easy - try filling up the float bowl to just under the overflow, then tilt it a few degrees to the left.  The fuel is now on an angle relative to the carb and gushes out the overflow.  The 400 has no fuel regulation, so I wonder if the pressure effects of a 500cc pony tank on the bench vs a full fuel tank on the bike are significant as well. 
Title: Re: Solution for float level adjustment vexation (?)
Post by: Rikugun on March 16, 2014, 09:58:18 AM
QuoteThe best I could do was to raise the supply tank about 6 ft above the needle valve
I've been told it's 1psi per 2.3' for reference. I have no idea how that compares to the pump or what the pump develops.

QuoteI wonder if the pressure effects of a 500cc pony tank on the bench vs a full fuel tank on the bike are significant as well. 
Provided we're talking gravity feed only, the elevated 500cc pony tank would provide more pressure than the installed fuel tank. The volume is of no consequence, only the free surface compared to the point of reference or in this case, needle valve.

Title: Re: Solution for float level adjustment vexation (?)
Post by: pullshocks on March 18, 2014, 12:03:25 AM
Well I have got my 2 floats set at 21 mm in the test jig.  We will see how that works out in use.  With the overflow tubes at 16.5 and 17.0, and a fuel level of 21 mm, I am not going to be surprised to see some fuel in the plastic tubing that drains the overflows.

I had a lot of trouble getting my front carb float to behave.  Over the course of many adjustments, the pivot hinge got a bit twisted, and started to bind up on the pivot pin.  Check this carefully.

It appears to me that depending on fore and aft and side to side lean angle, the floats are going to function differently and allow differing amounts of fuel in the bowl.  I believe the carbs must be designed to work within a range of such conditions. 

At the pressures in my test system,  the float compensated as I varied the rate of fuel draining out, maintaining the fuel level at the set point. 

The buoyancy of the float effectively closes the needle valve against the pressure in the test.

At higher RPM, presumably the fuel pump produces more pressure.  Whether that pressure overpowers the float, I don't know.  But the rate of fuel use would to up too, offsetting any such effect.

I figured out a way to test at higher pressures--get a big syringe.

But I am not pursuing that right now.  When I do, I will set up a fan to blow the gasoline vapors away.  Very unpleasant.

After working on this I believe the float buoyancy will hold against considerable pre
Title: Re: Solution for float level adjustment vexation (?)
Post by: Rikugun on March 18, 2014, 10:32:57 AM
Good deal  :) I don't doubt the float system will handle expected design pressure with some wiggle room. The pump has a built in bypass when pressure exceeds demand.

What often is the case may be an issue with debris fouling the needle. Even when the tank looks "clean" and a filter is used, tiny bits is all it takes. Also key seems to be striking a balance between prescribed bowl fuel level and overflow tube height. It seems there may have been some QC issues in this department.
Title: Re: Solution for float level adjustment vexation (?)
Post by: pullshocks on March 19, 2014, 12:48:18 AM
Well the 82 fuel pump has the bypass.  83 has the fuel return, so no bypass.

So the 83 pump has to have enough pressure designed in to force fuel back into the tank.  Presumably the float valve is spec'd to hold against that amount of pressure. (But as we all know, the designers made some questionable judgements on numerous design features.)

I am tempted to drill a whole in the carb body right at the 20 mm level and install a sight glass.


Title: Re: Solution for float level adjustment vexation (?)
Post by: Rikugun on March 19, 2014, 11:09:06 AM
Quote from: pullshocks on March 19, 2014, 12:48:18 AM
Well the 82 fuel pump has the bypass.  83 has the fuel return, so no bypass.
Yup, my bad. You did specifically mention these are '83 carbs. I'm not as familiar but both carbs use a "regulator" of sorts? The '82 loops excess gas via the regulator within the pump and the '83 regulator loops it back to the tank.

I suppose the regulator itself could be at fault but I tend to doubt it. I don't know what these pumps produce pressure wise but I suspect it's quite low - single digit psi perhaps? Pulse pumps are used on lawn equipment, snowmobiles, etc. (often with no bypass) and don't overwhelm the float system. Even carbureted cars using positive displacement mechanical fuel pumps seem to cope OK.
Title: Re: Solution for float level adjustment vexation (?)
Post by: The Prophet of Doom on March 19, 2014, 01:42:20 PM
Quote from: pullshocks on March 19, 2014, 12:48:18 AM
I am tempted to drill a whole in the carb body right at the 20 mm level and install a sight glass.
Excellent idea.  First person to do a carb sight glass mod wins a prize.
Title: Re: Solution for float level adjustment vexation (?)
Post by: fret not on March 20, 2014, 12:44:40 AM
And you're going to seal it with what?

It would be nice to have a float level adjustment from outside of the carbs.  How hard could that be? ;)
Title: Re: Solution for float level adjustment vexation (?)
Post by: pullshocks on March 20, 2014, 02:02:09 AM
Haven't figured that out yet.  Seems like it would have to be glass....

I don't know how the float could be made adjustable from the outside... it would sweet.

Another pipe dream of mine is to be able to extract and clean the pilot jet without removing the carbs.  In my experience that is the location most susceptible to plugging.  I don't think I have ever found any of the other jets plugged.
Title: Re: Solution for float level adjustment vexation (?)
Post by: fret not on March 20, 2014, 08:27:41 AM
The most reliable materials would be O ring or gasket, but either of those would require a mechanical means of holding it all in place.  There is probably some space age goo in a tube that would do the job but I don't know what that would be.  I wonder how resistant to petroleum based fuel the cyanoacrilates (super glue).  There is a variety of them, and one is a bit flexible. 
Title: Re: Solution for float level adjustment vexation (?)
Post by: pullshocks on March 22, 2014, 08:45:26 PM
Well, not out of the woods yet.

Reassembled the bike.  On the center stand, not side stand.  Tried to start.  Fuel came gushing out the overflows.

Took the tank and air box back off.   After trying a few things, I believe what is happening is the float is hanging up on the drain tube.  The foam is shaped to provide clearance around the tube, but I can see that the float is slightly angled in a way that would bring it closer to the tube. 

I hope I can bend the hinge to overcome this.
Title: Re: Solution for float level adjustment vexation (?)
Post by: pullshocks on April 07, 2014, 12:00:34 AM
Back in the float valve laboratory today.  Came up with a way to manipulate the float from outside the carburetor.  Ran a fine wire through the vent hole and tied it to the metal part of the float.  A gentle tug on the wire lifts the float.
With the carb reassembled you can hear the float moving as you  gently pull on the wire.  This verifies the float is moving inside the carb bowl.
Title: Re: Solution for float level adjustment vexation (?)
Post by: pullshocks on April 07, 2014, 12:25:52 AM
So that test ruled out the theory that overflow was caused by the floats jamming in the bowls.

With that established it was back to the transparent carb bowl simulator.

What I observed was sometimes the valve closed, sometimes it didn't.  Very inconsistent.  When it worked, the float seemed to develop its buoyancy with about 80% below the fuel level, 20% above the fuel level.   Other times, the float would be completely submerged without shutting off the fuel.

After changing to fresh fuel, adding an in line filter,  and blowing compressed air through the valve, performance became consistent.

Conclusion:  it does not take much contamination to prevent the float valve from closing.

I reassembled the bike.  No more overflow at idle------ YAYYYYY!

Bike runs but needs an idle adjust and synch.

Title: Re: Solution for float level adjustment vexation (?)
Post by: Rikugun on April 07, 2014, 08:11:58 AM
Quote from: pullshocks on April 07, 2014, 12:25:52 AM
Conclusion:  it does not take much contamination to prevent the float valve from closing.
No truer words have ever been spoken.....

Congrats and here's to continued leak-free times.  :D
Title: Re: Solution for float level adjustment vexation (?)
Post by: QBS on April 09, 2014, 12:15:52 AM
To clean debris out of a float needle valve: remove main fuel line from carb....attach short piece of separate gas line to carb......open carb drain screw.....tap side of float bowl to ensure float drop.....fill large syringe or reasonable facsimile with gasoline or carb cleaner........stick output end of syringe into short fuel line.....forcefully inject gas in syringe into gas line.....repeat several times.....reconnect main fuel line to carb.......close drain valve........turn petcock to PRI......check for leakage.
Title: Re: Solution for float level adjustment vexation (?)
Post by: pullshocks on April 09, 2014, 10:02:49 AM
Great suggestion, QBS.
Title: Re: Solution for float level adjustment vexation (?)
Post by: Jirik on April 13, 2014, 02:49:02 AM
Great topic, thanks for it.
What do you think about this advice https://www.youtube.com/watch?v=P0egD-1XWN4 (https://www.youtube.com/watch?v=P0egD-1XWN4)?
That is a little brutal for float valve spring, but it probably works.
Title: Re: Solution for float level adjustment vexation (?)
Post by: fret not on April 14, 2014, 12:58:37 AM
Yes, a bit brutal to the little spring.  It is toast, literally.  When most metal glows that bright is has no temper remaining.
Title: Re: Solution for float level adjustment vexation (?)
Post by: Rikugun on April 14, 2014, 08:26:23 AM
While I'm cleaning the carb bodies I soak all the brass bits in a small container of solvent. This generally will free up a sticky, varnish laden inlet needle damper. If rubber tipped, you may want to limit exposure but non rubber tipped can be soaked at length.