Is there a carburettor guru in the house?

[Andy 308GTB]

In the last few weeks I have rebuilt 4 Weber (40DNCF) Twin Choke Carbs
By rebuilt, I mean I have cleaned them inside and out, replaced the gaskets, accelerator pumps, needle valves etc.
They are now installed. The linkages and air bypass screws have been adjusted such that the air flow is consistent across each choke.

i started the car with the idle mixture screws 2 turns out and the car idles satisfactorily (but lean as anticipated).
I screwed the idle mixture screws out another turn and the 'spitting' from the chokes reduced. The car burbles away quite happily.
The idle speed screws are holding the car at a steady 1,000 rpm.

Here comes the but...
But, if I blip the throttle to, say, 2,000 rpm, the revs stay at 2,000 rpm for a long time, I mean 30 or 40 seconds before dropping back to idle.
I have a number of theories but I would like to hear the opinions of people who know about carbs.

Thanks
Andy

[fergus]

Andy

Hi. I would check:

Do the throttles snap shut when not connected to anything?

Does this happen if you disconnect the link/actuation rod and then blip the carbs by hand?

I think you may have something sticking/binding somewhere or a possible air leak. Also check: Bearings for the throttle spindles (gritting/grinding, or impeded by the leather washer)
Bind in the spindles per carb body

I'd recommend installing a lambda boss in each set of primaries as you can then install a data logging afr meter and jet / setup your carbs as optimally as you can without needing to go near a dyno

[Lotobear]

I would be thinking air leak somewhere - have you tried the 'trick' of spraying some WD40 or carb cleaner around the intake gaskets and spindles?

[Auntieroll]

There are some small return springs on the butterfly spindles which have a habit of failing ,
these tend to snag and can cause the symptoms the OP described.
HTH.

[Andy 308GTB]

Thanks for the responses. I am on top of a number of items already...
Apologies for the length of the reply.

Throttle spindle bearings - I packed these with grease and I am very certain that they are sealed off properly. I did have sealed bearings ready to go in but I decided to leave this job until the end of the driving season - I don't have the patience to make a mini bearing puller at the moment.

Butterflies closing - I dismantled all the springs and mechanisms and lubricated them. The butterflies snap shut (when the Carbs are on the workbench...)

Vacuum Leaks - painfully aware of the problems these cause.
On the workbench, I flatted the base of the carburretors so they seal against the inlet manifolds. I checked the inlet manifolds for signs of leakage around the gasket to the engine block but they seemed fine.
3 of the carbs I did not unscrew the butterflies and remove the spindles but only after I had learnt what a pain in the arse disassembling them is (and that there is little or no gain in doing so). I did find that the butteflies on the one I fiddled with weren't sealing off properly, I could see daylight though the choke... so I readjusted them until they sealed off the choke when closed.


Finding vacuum leaks...
Spraying WD40 or carb cleaner, I've even seen people on YouTube using butane gas... I guess this is a sound method. I imagine the revs would rise when using carb cleaner or butane gas. Would they rise with WD40 too? Either way I will try it tonight!

Removing linkages
The idle speed is set using the stop on the linkages. I think I would be able to set the idle on 2 of the carbs but the other 2 won't have stops. I'm sure there is a way around this and it seems the best way to eliminate the possibility of the linkages binding. I'll see what options I have this evening. (EDIT - I could remove the front to back linkage and test the carbs in pairs)

One more question...
The problem with the engine not returning to idle promptly. Is there any way this could be caused by the mixture being too rich or too lean? No idea how but it is something that troubles me

Gratuitous photo of what I am up against...

Edited by Andy 308GTB on Wednesday 8th May 15:39

[Lotobear]

..in theory the liquid sprays should interfere with the vac leak and result in a drop of revs and the gas should get drawn in resulting in a rise. Incidentally I tried all of these 'tricks' when trying to resolve the same issue but, whilst it was ultimately a vac leak, the spray trick did not find it. I guess it will depend on how severe the leak is?

I've no experience of downdraught carbs, only sidedraft Dellorto's but what I found was perished rubber mounts (Alfa) and insufficiently tightened flexible carbs mounts (Lotus TC) caused this hanging revs effect. I'm not sure if you have rubber or flexible mounts (as I understand it this feature is to stop the fuel foaming in the float chambers) but I would look at these first if you have

On the Lotus and Alfa I found that grabbing the carbs and pulling them up and down caused this effect as the mounts flexed and opened - on both cars the mounts were at fault and the hanging rev issue was resolved by fitting new Misab plates on the Lotus and solid mounts on the Alfa.

Best of luck - those buggers look complicated!!

PS: the balance mechanism on your 308 looks identical to the side drafts - not saying this is your problem but I found that you could get the butterflies out of sync quite easily between carbs at idle which would also result in what you describe. I'm probably telling you how to such eggs but my starting point (before balancing) was to back off the sync screw and push both butterflies right back so they are completely snapped closed, then screw the adjuster so it's just touching the lever. I found it's quite easy to have one butterfly slightly open at idle and the other fully closed and still have it run fine but then get a hanging idle. I then use a Morgan carbtune to fine tune the balance.

Edited by Lotobear on Wednesday 8th May 17:05

[Andy 308GTB]

Excuse my thinking out aloud, rambling posts...

Andy 308GTB said:

Removing linkages
The idle speed is set using the stop on the linkages. I think I would be able to set the idle on 2 of the carbs but the other 2 won't have stops. I'm sure there is a way around this and it seems the best way to eliminate the possibility of the linkages binding. I'll see what options I have this evening. (EDIT - I could remove the front to back linkage and test the carbs in pairs)

I removed the front to back linkages, so the 2 pairs of carbs (front and back) were independent of any outside interference.
But the problem persisted and the idle did not drop back to 1000rpm in a timely manner...

Lotobear said:

PS: the balance mechanism on your 308 looks identical to the side drafts - not saying this is your problem but I found that you could get the butterflies out of sync quite easily between carbs at idle which would also result in what you describe. I'm probably telling you how to such eggs but my starting point (before balancing) was to back off the sync screw and push both butterflies right back so they are completely snapped closed, then screw the adjuster so it's just touching the lever. I found it's quite easy to have one butterfly slightly open at idle and the other fully closed and still have it run fine but then get a hanging idle. I then use a Morgan carbtune to fine tune the balance.

This is an interesting point. The instructional pieces on tuning these carbs focus on getting the air flow equal across all chokes at idle (which I have done). But this assumes that all 4 chokes close perfectly. If one is letting a small amount of air through when fully closed, this process will simply open up the other 3 carbs to match. The fact that they are pulling the same amount of air through doesn't mean that the butterflies are fully closed - it could simply mean they are all open the same amount...


The bases of these carbs are some weird material that has some give it in (but doesn't look like rubber to me) and I am very confident that it is sealing off OK, especially after I flattened them off with very fine sandpaper. These bases are then compressed onto the intake manifold with 4 studs and 13mm bolts, so I think they are airtight.

I fear I will be taking these units off the car again and fiddling around with the butterflies. Joy.

[99hjhm]

One thing that works on these carbs is to have the holes in the sides of the idle jets at 45 degrees to the flow of air through the carb. Mark the holders at the top, then unscrew and realign the jets in the holders.

I realise that’s probably not your problem.

[fergus]

Andy 308GTB said:

Excuse my thinking out aloud, rambling posts...

Andy 308GTB said:

Removing linkages
The idle speed is set using the stop on the linkages. I think I would be able to set the idle on 2 of the carbs but the other 2 won't have stops. I'm sure there is a way around this and it seems the best way to eliminate the possibility of the linkages binding. I'll see what options I have this evening. (EDIT - I could remove the front to back linkage and test the carbs in pairs)

I removed the front to back linkages, so the 2 pairs of carbs (front and back) were independent of any outside interference.
But the problem persisted and the idle did not drop back to 1000rpm in a timely manner...

Lotobear said:

PS: the balance mechanism on your 308 looks identical to the side drafts - not saying this is your problem but I found that you could get the butterflies out of sync quite easily between carbs at idle which would also result in what you describe. I'm probably telling you how to such eggs but my starting point (before balancing) was to back off the sync screw and push both butterflies right back so they are completely snapped closed, then screw the adjuster so it's just touching the lever. I found it's quite easy to have one butterfly slightly open at idle and the other fully closed and still have it run fine but then get a hanging idle. I then use a Morgan carbtune to fine tune the balance.

This is an interesting point. The instructional pieces on tuning these carbs focus on getting the air flow equal across all chokes at idle (which I have done). But this assumes that all 4 chokes close perfectly. If one is letting a small amount of air through when fully closed, this process will simply open up the other 3 carbs to match. The fact that they are pulling the same amount of air through doesn't mean that the butterflies are fully closed - it could simply mean they are all open the same amount...


The bases of these carbs are some weird material that has some give it in (but doesn't look like rubber to me) and I am very confident that it is sealing off OK, especially after I flattened them off with very fine sandpaper. These bases are then compressed onto the intake manifold with 4 studs and 13mm bolts, so I think they are airtight.

I fear I will be taking these units off the car again and fiddling around with the butterflies. Joy.

Are the carbs mounted on misab plates (i.e. rubber o-rings "bonded" on to a mounting plate)? Are the mounting nuts tight?
All it takes is a slight misbalancing between carbs to allow one throttle plate to be open by a sliver and you have an "air leak" which will cause the revs to fall slowly. If you manually "close" the carbs, does the problem persist?

If so, it could be the ignition (if you have a std distributor fitted the advance weights may be sticking giving the car a lot more advance than it needs under idle conditions. This will raise the revs a lot). Do the revs eventually fall? If so, I would get a timing light on the front pulley and check the advance whilst the revs are falling. I would also manually close the throttle plates and perform this exercise. This way you can rule out any ign related issue.

Did you get a chance to put an airflow meter (or check the vacuum) across the carbs to check they are 100% syncd?

Are the carbs operated via rods or via cables? Do these both mechanically allow the throttle plates to close 100%

[Andy 308GTB]

Hi Fergus, thanks for your input.

The bases of the carb are a solid piece of material, which covers the entire base of the carb, it does have some give it in. The bases of the carb are secured to the inlet manifold with 4 studs that take 13mm nuts - so they are firmly secured. There were some impressions on the base (from being in position for nearly 40 years) - I flattened these off, to get the best seal. I would upload a photo but I am at work - but I will do so when I'm home tonight. I don't think the bases are the issue.

fergus said:

All it takes is a slight misbalancing between carbs to allow one throttle plate to be open by a sliver and you have an "air leak" which will cause the revs to fall slowly. If you manually "close" the carbs, does the problem persist?

I think this is the likeliest cause. The butterflies do snap shut on each carb and the I confirmed last night that the linkages don't bind.
When I removed the carbs, I did adjust one of the butterflies to get it to seal off better, I think I may need to do have another go on this carb and check the others more carefully.

The ignition was rebuilt several years ago (balance weights and hall effect sensors). This car has NOT been laid up for years, it was running OK towards the end of last year, so I'm am reluctant to blame/question the ignition system.

I did sync the 8 chokes up using an airflow meter - they are all pulling the same amount of air through.

My task this evening is to enrich the mixture beyond the recommended 2 1/2 or 3 1/2 turns that is recommended. When I disassembled the carbs the idle mixture screws were way out (5 or 6 turns) which seems crazy. Before I started this venture, my feeling was that the car was running rich and I wanted to set it up to be 'stock', hence trying to get it work with around 3 turns.

I don't understand how the fuel/air mix being lean could impact the speed of the return to idle!

[Andy 308GTB]

It looks like this is now fixed...

A good friend on FerrariChat came up with this:

"If the mixture is too lean and you compensate idle speed with opening the throttles more, the transition/progression circuit comes into play. The bigger airflow draws fuel through the progression circuit and this affects idle speed coming down. It may not be enough fuel flow through the progession circuit to keep idle speed permanently high, but enough to slow things down. This is the answer I just got from my Weber-specialist.
Next step. Close the butterflies more and screw out the mixture screws. I'm curious what happens"

So this evening, I screwed out all the idle mixture screws 5 turns (i.e. way beyond the suggestion of around 2 or 3 turns), screwed out the idle speed screws so they were not in play (i.e. and the butterflies were completely closed).
I had to use a 1/2 turn of the idle speed screws to maintain the idle but all went well and the when blipped the revs drop straight back to idle.

The game now is to lean off the mixture whilst not having to use more turns of the idle speeds screws.

Thanks all for you support & input - it is much appreciated.

Edited by Andy 308GTB on Thursday 9th May 19:38

Edited by Andy 308GTB on Thursday 9th May 19:39

[fergus]

Andy 308GTB said:

It looks like this is now fixed...

A good friend on FerrariChat came up with this:

"If the mixture is too lean and you compensate idle speed with opening the throttles more, the transition/progression circuit comes into play. The bigger airflow draws fuel through the progression circuit and this affects idle speed coming down. It may not be enough fuel flow through the progession circuit to keep idle speed permanently high, but enough to slow things down. This is the answer I just got from my Weber-specialist.
Next step. Close the butterflies more and screw out the mixture screws. I'm curious what happens"

So this evening, I screwed out all the idle mixture screws 5 turns (i.e. way beyond the suggestion of around 2 or 3 turns), screwed out the idle speed screws so they were not in play (i.e. and the butterflies were completely closed).
I had to use a 1/2 turn of the idle speed screws to maintain the idle but all went well and the when blipped the revs drop straight back to idle.

The game now is to lean off the mixture whilst not having to use more turns of the idle speeds screws.

Thanks all for you support & input - it is much appreciated.

Edited by Andy 308GTB on Thursday 9th May 19:38

Edited by Andy 308GTB on Thursday 9th May 19:39

Webers used a variety of profiles on the idle mixture screw over the years. 5 turns out, even on needles with a very fine taper suggests you most likely need richer idle jets (go up on the first number (fuel hole size), but not on the "F" number (air bleed)). This will also affect your progression, as the progression circuit is feed from the supply from the idle jets.

You *may* encounter sluggish low-mid range (2-3k rpm) running with your current setup.

[Andy 308GTB]

fergus said:

Webers used a variety of profiles on the idle mixture screw over the years. 5 turns out, even on needles with a very fine taper suggests you most likely need richer idle jets (go up on the first number (fuel hole size), but not on the "F" number (air bleed)). This will also affect your progression, as the progression circuit is feed from the supply from the idle jets.

You *may* encounter sluggish low-mid range (2-3k rpm) running with your current setup.

Thanks Fergus, I believe the correct adjustment range is around the 2 or 3 turns mark.
I haven't tuned the car yet, I'll do that this weekend and I will attempt to lean off the mixture whilst not opening up the throttles any further.
But I very much doubt that I'll get down to the 2 or 3 turns marks - as my problems, detailed in previous posts, were with the idle mixture screws screwed out up to 3 1/2 turns.

Let me see how I get on - I might be back to ask some questions re the jetting.

[Lotobear]

I've found that the recommended 'turns' out on the idle mixture screws can be extremely variable.

One recognised way is to keep adjusting the mixture screw by 1/4 turn, leaving a good while between turns, until you achieve the highest manifold vaccum on that choke (a 'manometer' type balancer is great for this such as Morgan Carbtune as you can see excatly what is happening) as that represents the correct burn.

Another way is to use a Gunson Colourtune but I've had mixed results with this.

I usually just keep adjusting until the revs stop rising/start to fall back. I've been very surprised at how many turns this can end up at - often way more that what is recommended in the 'book'

[fergus]

Lotobear said:

I've found that the recommended 'turns' out on the idle mixture screws can be extremely variable.

One recognised way is to keep adjusting the mixture screw by 1/4 turn, leaving a good while between turns, until you achieve the highest manifold vaccum on that choke (a 'manometer' type balancer is great for this such as Morgan Carbtune as you can see excatly what is happening) as that represents the correct burn.

Another way is to use a Gunson Colourtune but I've had mixed results with this.

I usually just keep adjusting until the revs stop rising/start to fall back. I've been very surprised at how many turns this can end up at - often way more that what is recommended in the 'book'

The only real way to do this is with a (wideband) logged AFR meter. "Flow" measured in vacuum terms can be misleading at times as the engine is only a "pump" pulling in air and fuel. Anything measuring flow won't account for the AFR of the mixture. Correct jetting without an AFR meter is at best guesswork (IMHO).

[Lotobear]

That's a very good point but I guess not every home mechanic has access to such equipment.

That said, I did have a Lambda boss fitted into the exhaust on my Elan which never got used....now there's a thought!

[fergus]

Lotobear said:

That's a very good point but I guess not every home mechanic has access to such equipment.

That said, I did have a Lambda boss fitted into the exhaust on my Elan which never got used....now there's a thought!

http://www.zeitronix.com/Products/zt2/zt2.shtml + a loptop and you're off and running!

[GreenV8S]

It's common to use a tailpipe mount for that rather than add a boss to the headers.

[fergus]

GreenV8S said:

It's common to use a tailpipe mount for that rather than add a boss to the headers.

I was of the impression that most people add their boss for AFR reading to either the primaries or secondaries to ensure a decent signal?

[GreenV8S]

It's better, but a tailpipe reading can be adequate and obviously a lot easier for people who do not plan to use WBO2 long term.