Fitting a Carter 4070 pump

[nomad]

Oh, and the "Gold" Facet in its Bendix form at least make's a, not objectionable, clicking sound to let you know its working.

Kurt.

[nomad]

Just read up on the Facet that awatkins and myself are talking about and Facet claims it's new electronic switching is more reliable than the previous points. Don't know how they could beat that old Bendix!!

Kurt.

[simonknee]

Pumps been in for a week and here's my notes so far.

The Facet is not to blame for it's loud noise just the noise signature. It is something to do with my larger bore fuel system. I say this because the Carter also takes 5-10 minutes before it settles down to it's quietest. However I prefer it's whine to the Fact pock - so all good there.

I bought a little 1-15psi fuel pressure gauge from Burton Power (very reasonable). So can verify that the Carter puts out 4psi on the nose. This is measuring with the carbs disconnected and engine not running. So this is the highest pressure that could be measured. I don't have a junction piece to measure the pressure in situ, yet. Does anybody know where you measure the pressure and whether the engine should be running or not.

Now the world and his wife has an opinion on fuel pressure (and pumps) with Webers but non of it seems backed up with real reasoning. Yes too much will over-power the needle valves but how much? Have people actually tested or is it received wisdom (or did they have worn needle valves!)? Webers period literature states 1.5 to 2.5psi. and t'ineterweb repeats 3.5psi max but I've not found out why. There is also some rubbish out there - for instance "one carb needs 3.5psi so twin carb setup will need 7psi" - I'm no fluid mechanic but that is obvious rot.

To my mind if you can set the float level (at 25mm of course) and it doesn't creep up higher then your needle valve is operating correctly - am I missing something here? Also if you don't end up with fuel starvation at WOT then you have enough pressure, no? It's a very simple mechanical system. I am going to visit Keith's Yahoo group and see what he thinks as I think he has more Weber DNA than human by now.

Anyway I can't find my old Sytec fuel regulator in the garage right now so that's settled it for the time being.

cheers

Simon

[awatkins]

Simon --

You might consider that if the fuel pressure is such that the closed valve is just barely able to resist fuel inflow under static conditions (idling in your driveway) it might not be able to on the road at all temperatures, during bumps, varying vibration levels, age, cornering, braking, etc, and that the results would be fuel overflow and possibly a fire. Of course only you can decide whether that risk is offset by the trouble to install a pressure regulator

Yes, a static drawing of that part of the fuel system is intellectually simple but its behavior under short termr dynamic conditions and over the long term with component wear and accumulation of residue is not simple at all.

Also, it should take only a few seconds to fill even empty float bowls, so 5-10 minutes for the sound to stabilize doesn't make sense to me.

[simonknee]

Thanks for the thoughts...

I don't think it is the filling of the bowls that is taking the time. Besides there is no reason for them to be empty since we are talking only a few hours since the car last ran. This is easy to verify, so I will. Also I have no running issues with the Carter or the Facet before it. Just an increased noise for this extended period of time. I think it is getting some kind of air-lock out of the pipe or something. Odd I know. It is probably related to the fact that the tank only has a 1/4" pipe in it which then goes to my 3/8", 6AN connected, fuel system.

Remember that the initial problem was just about reducing the noise. My carbs run great and I have a wideband O2 sensor to prove it!

I have spent some time today trying to understand what kind of a system the float needle valve actually is and how it is affected by volume and pressure. Not got the full picture yet however it is basic fluid mechanics so if somebody knows what it is electrically analogous to then I could probably understand it!

Simon

[awatkins]

If your pump is connected a to the tank by anything solid that will explain the noise, and that tank will act as a wonderful acoustical radiator.. To make it quiet you would have to use at least a few inches of flexible tubing on each side, and rubber mount it the way Carter intends.

Electrical analog to a float valve? If you mean the valve itself, but not the float, it's a really non-linear variable resistor of infinite value. (where I = flow and V = pressure). But I don't see the utility of that component when your variable resistor is controlled by a float sitting on a tub of gasoline moving around in three-space. If you can make an electrical model of that you deserve a nobel prize.

[simonknee]

Yep, I have a few inches of rubber hose to the tank. Recommended by Andy the tank maker to prevent stress failures. The is no rubber tubing on the other side of the pump - I want as little of the nasty stuff in my boot as possible. But even if it there was no rubber how would this explain the 5-10 mins of gradually decreasing noise?.

Well the tub of gasoline is a capacitor, yes/no? The float is measuring the voltage stored and adjusting the variable resistor to allow more/some current to flow into the capacitor until the voltage is up to a specified set point?

The slosh is 3 dimensional but the float only goes up or down. Fuel pressure doesn't affect the slosh in the bowl. So why worry about the pressure as long as it is not enough to actually push the float "underwater"?

[rgh0]

You need sufficient pressure delivered by the pump to get the required flow rate down the the line as the valve opens with a slightly lower fuel level in the carb bowls.

To low pressure and you will get not enough flow or the valve has to open to far to get enough flow and the fuel level drops and the carbs go lean.

To much pressure and the fuel level goes high until the floats are more under the fuel level to generate enough force on the valve to close it. If much to much pressure and the floats cant close the valve and the carb bowls overflow.

The 2 to 4 psi range seems to work for Webers with the standard Elan set setup

cheers
Rohan

[awatkins]

I didn't say the rigid tubing would explain the decay of the noise. I was explaining why it's loud.

As for the electrical analogy: way off topic.

[simonknee]

Hi Rohan,

Thanks for your input. This is a quest for me to better understand the mechanics of this part of the system. i.e. why does 2-4psi "seem to work well". And of course this leads me to ask should I choose 2 or 3 or 4, how do you determine the optimum pressure for a given setup? I understand the fault modes of grossly under or grossly over pressuring. So let me ask the question a different way.

As awatkins rightly said the static mode is easy to understand. So lets consider the running scenario. Say the car is at a fast idle. And lets imagine we have a pump plus a regulator set to 3psi. In this situation set the float to give a fuel level of 25mm. This has set up an equilibrium where the float is keeping the needle valve open enough to allow as much fuel in as the engine draws out. Now without changing any other variables alter the regulator to first 2psi then to 4psi. What happens? Does the float compensate and bring the level back to 25mm? Or do I get lower level for 2psi and higher for 4psi?

I am 90% certain you are going to say the level does change as the pressure on the needle valve is pushing the float slightly lower in the fuel for a given input flow and so more fuel is in the bowl to push the float/needle back to the equilibrium point. Am I right or wrong? Is it significant, i.e. greater than 1mm of level change?

( BTW electrical analogy is completely "on topic" for a fluid mechanics discussion. Perhaps you meant I was way off the mark with my attempts to at it, awatkins, )

[rgh0]

Hi Simon
Have you got involved in process control at all via electronics? If so the float valve is a simple proportional controller.

The set point is the nominal 25mm fuel level pushing against the float to close the valve. To open the valve the fuel level will have to drop. The higher the fuel consumption required by the engine the further the fuel level has to drop to open the valve further to match the required fuel flow.

Simple proportional controllers require an offset from set point for the controlled action to happen ... in this case fuel flow. The more the fuel flow the larger the offset from the nominal 25mm set point

The float control valve however has a "high gain" so a small level movement in the fuel results in a large fuel flow increase. The change in fuel level over the engine operating range is small enough not to worry about provided the fuel pressure is in the 2 to 4 PSI range over the range of fuel demand it appears.

The fuel pressure at the carbs will normally drop as flow increases as most pumps have a pressure curve that drops with flow and also you get increasing pressure drop in the line to the carbs with flow. The pressure drop increase with flow will increase the level offset required to achieve the needed flow to match the engine use. However again provided the fuel pressure stays in the 2 to 4 PSI range you should not have a problem with the standard Elan setup.


cheers
Rohan

[simonknee]

Hi Rohan,

That makes sense and the fact that the float system itself will alter flow and hence the measured pressure goes some way to explain why a working pressure range is considered appropriate. This leads me back to an earlier question of where should you attach a pressure gauge to look for the 2-4psi? I don't have a take off point yet so all I could measure is the Carter 4070 straight into the guage - 4psi - this I assume indicates the maximum pressure it can do.

Also can you give me an answer to what would changing the regulator setting from 2-4psi do to the fuel level with the engine running and fuel flowing i.e. if I had a regulator and changed the setting on it do I need to reset my float level?

I found my fuel reg this morning so I am nearly in a position to provide my own answers. I don't trust the calibrated markings on it (mainly because ten years ago I fiddled with the grub screw ) How would you set a regulator? Again this takes me back to where should I fit the gauge to get useful readings?

Simon

[awatkins]

Simon -- No, I meant a debate over whether a useful electrical analog could be created and how it could be done would bore everyone else to tears and isn't going to help someone looking for information about "Fitting a Carter 4070 pump" to a "Lotus Elan."

So back on the subject of noise reduction, I see your pump bracket is mounted straight to the body; there is probably flexing in that area comparable to that of the carter isolation mounts so I suspect your noise is being transmitted through the GFRP material, as well as the hard line running forward. Also, when I was suggesting flexible lines I was thinking of typical "aircraft" stainless steel armored line in longer lengths than you have, carefully dressed so as not to be under static tension. Here it would be advantageous to use smaller hose than 3/8" because it's more flexible, and your maximum fuel flow rate does not demand 3/8" hose especially over such a short distance. To do this properly you would need to arrange a proper flare or compression coupling to the tank. I agree with your reservations about hose-clamped rubber tubing, not only because the hose-clamped junction is not reliable, but also because you need to make sure the rubber line is rated for fuel use; anything else will degrade over time with the constant contact with gasoline and its additives. Using the teflon lined variation of "aircraft hose" and AN fittings throughout solves all these problems. As was said earlier, a Facet "interruptor" or similar piston-operated pump mounted in this way makes muted thumps with each cycle that are barely audible.

Regarding the regulator location, the closer to the carburetors the better, although with 3/8" lines it won't really matter. But if it's closer to the carbs the dynamic response of the regulator with changes in load is closer to ideal.

Regarding where to tap in for a pressure measurement, since that's an essentially static measurement it doesn't matter as long as it's somewhere between the regulator and carburetors. There would be no significant static pressure drop along that path under any regime you would be measuring.

[simonknee]

I thought they were tears of happiness that people wept when I give them an explanation... boredom you say, I suppose it's possible?

I do have sheet of rubber betwixt body and bracket - I am going to up-rate this to neoprene (a mouse mat).
I would use rubber standoffs but there isn't enough room for the spare tire after that.

I did plan to have Andy (http://axminsterspecialistpanels.com/) swap the 1/4" pickup for a 3/8" and weld a 6AN onto the tank. However as you say this will get noisy. So a short run of something flexy would help in any scenario.

What I really need is just a few inches of SAE 30 R9 fuel hose. It is difficult to get unless you trust the ebayers that they are really selling is R9 spec. Funnily enough I don't!

Thanks for the other tips, make sense.

Still what I cannot figure is why both Facet and Carter take 5 minutes to "settle down" in my system noise wise. I suspect any pump would behave the same and it is something odd in my set up causing this.

[oldchieft]

Still what I cannot figure is why both Facet and Carter take 5 minutes to "settle down" in my system noise wise. I suspect any pump would behave the same and it is something odd in my set up causing this.

My guess would be you may be drawing air into the system, to get a good lead on whats going on, fit some clear hose at the carburettor inlet.

Then you will know what is happening in the 5 minutes in question.

Jon the Chief