Lotus Excel DHLA carburettors; float and fuel level setting

Well... if you have checked out my other blogs then you will know that I did get the motor running- but as it runs fuel drips into the Venturis, and puddles at the front of the air intakes and is very visible as I haven't fitted the filter box (thank goodness or I might not have noticed this!).
 From my previous post...
One problem I do note is that while running fuel can be seen dripping  into the Venturi... I'm not sure if this is just because I have been overplaying the throttle in an attempt to get the things started or if there is a problem with the jets seating, but the fuel collects in the intakes.

Engine idling- fuel collecting at carb intakes- all 4 show this.

Obviously I can't run it like this and as both carbs are affected I am assuming it was something to do with the rebuild. Clearly I'm not aware of having done anything wrong (or I wouldn't have done it) so trouble shooting leads me to consider everything.

I can say however that as a problem, this does seem to be not exactly unheard of and several forums have entries similar to this.

In general DHLA carbs may well drip from the acceleration jets at idle (especially if springs badly adjusted)  or from progression holes or other places if the float chamber overflows. One report suggests that an internal lead plug between the idle-air inlet gallery and fuel feed to the main jet can develop a leak and allow fuel to pass. I think this is a manufactured part of the carb and to be frank I don't really know where it is so I'm hoping it isn't my problem! Key in diagnosing which particular problem you have seems to be first and foremost- where is it leaking from and secondly under what conditions does it leak?

In my case I can say that just turning on the ign and letting the pump pressurise doesn't start a leak. Nor does using the acceleration jets a few times- they squirt and then stop- no dripping.

Acceleration pump pushes fuel through this brass jet at the rear of the venturi. No sign of excess dripping in my case.

The jets should work so that they are operated only by a sudden throttle opening, gradual acceleration movement is absorbed by the rod spring such that the acceleration pump arm isn't moved. Sudden movements compress the spring and it's the spring rather than the rod which operates the pump lever. I checked this adjustment and found it to be correct again reinforcing the possibility that the pump jets were not the culprit.

However, start running the car and pretty soon you can see fuel swilling around in the Venturis. I used my borescope to look inside the carbs while the motor was running- obviously you MUST NOT do this by eye- a backfire could cause you serious damage! The acceleration jets do not seem to be the culprit, but instead the fuel seems to be coming from the central tube of the venturi. The Venturis connect to channels in the carb via the wing section support strut that holds the central tube and this seems to be the source of the leakage.

Fuel could be seen dripping into the central tube of the venturi through this feeder arm 

Obvious explanations for this problem include too high a fuel pressure, too high a fuel level in the carb due to incorrect float setting or a leaking needle valve.

I have fitted a new pump but this is the correct pump designed to deliver around 3psi. I don't have a fuel gauge or a regulator but I have a gauge on order and will check on arrival. If necessary I can fit a low pressure regulator but as the pump is new and the same type (SU AZX1307EN) is in use successfully by others even without a regulator....Well my low pressure gauge (fuel and vacuum_ duly arrived- I get around 3.2 psi at the carbs which is just about ideal!

Gauge connected to carb fuel line... app. 3.2 psi

This pretty much eliminates fuel pressure as the problem and as I have fitted a new needle valve and seat body this can be eliminated too.

This leaves too high a fuel level which might allow excess to overflow.  As may see from the video, fuel accumulates in all carb throats, but in decreasing amounts from rear to front. As my car is parked on a sloping drive with the front highest this might also indicate something is amiss with fuel levels. At this stage I was pretty confident of my float heights having just checked them when I rebuilt the carbs... (Oh the folly of misplaced confidence! See on). However, one interesting post points out that the float height is only an indirect and approximate indicator of what really matters- which is fuel level in the float chamber. Even if the floats are perfectly set, the fuel level can still be off so its recommended to check this too.

In my motorcycle experience the best way of doing this is to connect the float chamber drain outlet to a clear tube to use as a manometer. Sadly no such drain holes in the DHLA and anyway this simple approach wouldn't work in a pressurised system! My twopenny-worth: considering these carbs are so advanced (and expensive), if fuel height isn't set by float height then it would have been nice to build in a way of checking it easily maybe a manometer tube with openings above and below fuel level?....?

Anyway, guided by that post and the ever-helpful (if occasionally tetchy) Tim Engel who once again provided a wealth of detailed information, I have now gathered that there are two methods for checking actual fuel level- and problems with both. The easiest is with "floats in" but there is no accepted value for that determination; only for the second which I'm calling the "Floats out" method and is taken from the sources above .... and I send my apologies to Tim ... I guess you can only ask so many daft questions???

I will describe my attempts below. If this makes it seem easy and straightforward then I can only apologise. Its fiddly, repetitive and tricky (almost random) to get right.

1. Floats in method (emulsion tube level)

The simplest method is to check the level of fuel actually reached in the carb float chamber whilst everything is both present and in position - trouble is although its the simplest method, there is actually no accepted figure for checking it this way.

First turn on the ignition for at least 30s to allow the pump to pressurise the fuel until it reaches its level. When it stops ticking, turn it off and wait a minute or so. Then remove the small plastic carb cover and its gasket.

Once off, remove one of the emulsion tubes for the main jet. These holes will be filled with fuel to the same height as the float chamber. I removed the "outer" tubes in each case (ie foremost on front carb and and hindmost on the rear one).

removing hindmost emulsion tube for the rear carb.

The recommendation is to then measure the fuel height using the extending (depth) end of some digital calipers... Good luck with that! You will find firstly that the base of the caliper won't fit onto the jet mounting boss so it wont dip vertically into the hole, and secondly that you cannot tell/see where the fuel level actually is anyway.   I tried to solve this by dipping the tube recess using a straightened plastic cable tie. I'm not saying this method is recommended, accurate or preferable- its just that this is what I did and these blogs are primarily a record of that. Anyway the tie will penetrate to the base of the float chamber and I found this to be 80-81 mm in depth. It's hard to be accurate because you can't look horizontally at the jet block- you have to look downwards at an angle which introduces parallax errors. I used a datum mark at 80mm from the pointed end of the tie which is close to the jet block height and all my measurements are taken from this 80 mm datum mark.

In order to detect the fuel level I coloured the cable tie (non- serrated side) with a fine point blue marker and repeated the dip, letting it sit in the fuel for 20 sec or so.

Cable tie sitting in tube hole

It is hard to see but the fuel will partially dissolve, and therefore fade, the coloured ink on the tie and its then possible to measure the  the distance from the datum to the fade line. In my case this was 15-17mm. If you have trouble finding the fade line then you can wipe the tie with a tissue from the top downwards.

Fade line in plastic tie- measure from the datum line (top of blue to the fade point).

The big trouble with this is that Dellorto do not publish a figure for the required level here and I'm not aware of any correlations between either float height or the floats-out fuel level and this level. The second method (see below) should show a fuel level of 27mm (again, empirically determined and not an official Dellorto figure) - however this is determined after the floats are removed and this will obviously lower the level of fuel in the bowl below that encountered in operation. This means that the level to be expected in the tube hole will be higher than that determined after the floats are removed (i.e. the depth value determined by dipping will be lower because you have to dip less far to reach a higher fuel level)... but its not clear how much higher! In Weber carbs Tim Engel tells me this difference is around 2mm, but there is no reason why this should be similar for Dellorto.

2. Floats-out method

In order to measure fuel level in the float chamber, the float chamber lid has to be removed. This presents two difficulties- firstly the floats will be removed lowering the fuel level from that actually encountered in a fully assembled and working carb. Secondly there will be some fuel leakage from the needle valve as it will open during the top removal process. This latter concern means that the top has to be removed smoothly and rapidly!

Remove two of the 4 top cover retaining screws

Remove two of the top screws. I didn't remove enough stuff and I should have disconnected hoses and cables as well as they obstruct top removal.

* Experience led to modifications of this method so see end of this entry for a step by step account of how I do it now!

You will need to detach the cold start cable and choke cable linkage from the carbs beforehand and get them out of the way. Turn the ignition on and allow the float chambers to fill for about 30s. Turn the power off and wait about a minute to let the pressure subside. You also need to remove the fuel hoses which are stiff and restrictive (wrap the ends in tissue or direct them into a container as you undo them since these may be under pressure). Then, whilst pressing down evenly on the choke tower and lid to stop it springing up and opening the needle valve, remove the remaining two screws. Finally, smoothly and rapidly remove the lid so as to avoid excess drippage from the needle valve which will add extra fuel to the bowl. You will need to make sure that you are also holding the gaskets around the base of the detachable top with your fingertips so that it comes up with the top and doesn't get caught or torn - as I now know from experience.  I didn't appreciate that you needed to remove all these items and so got the top tangled in the fuel pipe and tore the float chamber gasket. This also slowed me down and so led to fuel drippage into the bowl. I ordered a few extra gaskets as I suspect this may be a common problem! If you remove the fuel hoses then its relatively simple to remove the carb top, but its annoying to have to keep dis- and re- connecting them to check on the effect of any adjustments. Further, constantly disconnecting the hoses introduces air to the fuel lines this can affect the level achieved in the 30 sec pressurisation.

torn float chamber gasket.

However, once the top is off it is possible to measure the fuel depth using the depth measure probe of a pair of calipers: if the base of the calipers is rested on the jet block in the notch (you can do this now the top is off), then the tip of the depth bar should just touch the fuel at 27mm. Make sure you measure in the notch of the jet block (ie in the centre of the carb) and this should give some protection against the effect of slightly sloping ground.

Fuel levels in chamber now accessible- test fuel level from centre of jet block to fuel. The notch is visible beside the one remaining emulsion tube above the hose. You can also see the effect of the slope in my drive, raising the level on the left and decreasing it on right. Measuring in the jet block notch averages this out to some extent.

I found my fuel level was rather high at around 25mm- although this could be an over estimate because the level was raised by the extra fuel that dripped in during my ham-fisted removal of the float chamber lid. This was the first real discovery and obviously suggested that float height will need adjusting.

This is where I realised my error! Reviewing my blog (I knew these would be useful one day) showed that although I had checked the float height, at that time I had believed the correct height to be 12mm.  I have 8.5g floats (no longer available as replacements) as specified by Lotus and for which the float height as I now realise, should be 15mm in order to generate a fuel level (floats out) of 27mm. Clearly my floats are set too close to the lid- or too high which will in turn lead to too high a fuel level. (NB. Float height setting does differ with float weight as heavier floats sit lower and displace more fuel). Float level is adjusted by holding the carb top almost vertically until the float arm touches and seats the needle valve but does not depress the spring loaded tip.

Consequently the floats need to be adjusted. Most descriptions speak knowledgeably of "bending the tab" in order to adjust float height. This refers to the bifurcated tab that retains the needle valve. Access is tricky without removing the floats but the float height can be increased fairly easily inserting and twisting a screwdriver to lever this tab away from the float arms.

Using a screwdriver to twist and lever the small tab away from the float arms- ie towards the carb top and thus increase measured float height. Be careful not to damage the needle and be sure to bend both sides equally.

Adjusted float height- holding the carb top vertically and measuring from the lowest point of the float (closest to carb top) and gasket; float height is now  the recommended 15mm.

Of course if you do need to adjust the floats then you will need to remove the carb top, and the "floats out" method isn't such a bother. I was able to reset both floats to 15mm as shown. In order to test  the difference any adjustment has made, you need to bail out some fuel so that the pump has to replace it to the new level and then refit the carb tops and re-pressurise the fuel as before.- You don't need to refit the cold start system. I could then measure the emulsion tube fuel height and found this was now lowered to 20mm below my datum level. I also measured the fuel height by the floats out method; the level was now a little low at 30mm. I measured this several times and found that the difference between the two methods is usually of the order of  8-10 mm.

A low fuel height could lead to weakened mixture, fuel starvation and impaired performance, but too high a level could lead to fuel wastage and possibly fire. I also think that there isnt much leeway in terms of overly high fuel levels and overflow can start at only 1mm over the top (ie fuel level floats out 26mm). I think I prefer to err on the low side, so I left my carbs adjusted to 30mm fuel level (15mm float height) and I will see what further adjustments are necessary if and when I get the car running. Sadly I can't check to see if this procedure has cured the leakage on running the motor that got me started on this whole procedure. At present I am halfway through changing the pedal box so all controls and dashboard electrics are currently disconnected. I will need to fix the pedals and dash before I can check running.

I present below the method I arrived at finally after several attempts to address this issue:

Floats out Method
1. Check battery is charged, park car as level as possible.
2. Detach the cold start cable and remove choke cable linkage from the carbs.
3. Remove two carb top screws diagonally opposite each other- I usually remove the awkward ones such as the ID tags or choke bracket attachment.
4. Turn the ignition on and allow the float chambers to fill for about 30-60s. If you see air in the lines loosen the unions and let it bleed out This will almost certainly be needed if the lines have been disconnected/reconnected. If possible crank or run the motor to let levels settle.
5. Turn the power off and loosen the fuel union on the front carb (its easiest) to release pressure. Collect fuel spill in an aerosol cap for re-use.
6. When pressure is relieved remove the fuel hose completely- if checking front carb remove only this hose but remove both if checking the rear carb.
7. Whilst pressing down evenly on the choke tower and lid to stop it springing up remove the remaining two screws.
8. Grip with your fingers around the edge of the carb top, allow your fingers to also grab the gasket and then smoothly and rapidly remove the lid and gasket together.
9. Measure the fuel depth using the depth measure probe of a pair of calipers. Rest the base of the calipers on the jet block and measure  down the notch parallel to the pedestal but about 1mm away until the tip of the depth bar just touches the fuel surface. Make sure you measure in the notch of the jet block (ie in the centre of the carb) and this should give some protection against the effect of slightly sloping ground. Should be 27mm.
10. Adjust float closing (15mm) and droop (25mm) heights to aim for 27 mm fuel depth above and then refit the top using only two diagonal screws. Reconnect hoses and repeat to check effect. Remember and remove any air locks.