Showing posts with label spindle. Show all posts
Showing posts with label spindle. Show all posts

Wednesday, 10 May 2017

Rebushing the MGB SU carburettor HIF4

Rebushing (HIF4)

Note- There is a lot of confusion out there when it comes to this job. This is understandable because the MGB used different carbs at different times and they dealt with the problem of throttle spindle mounting in different ways. If you just want to follow what I did with the Hif4s just skip down to the section "Rebushing". If you are interested here is the result of my digging on the subject:

Background
The throttle spindle is a tricky compromise in carburettor design. Ideally, all air going into the carb should be regulated by movement of the throttle butterfly- but this has to be mounted on a spindle and thus there is always the possibility that air could be drawn in along the line of the spindle shaft, between spindle and its bore regardless of the butterfly position. Secondly the butterfly opens by twisting and allowing more air in above and below it as it opens. The centre of the butterfly should remain in smooth contact with the Venturi wall at all times and when closed should form a close seal against the Venturi wall. Consequently the spindle entry point should be both tight around the spindle shaft and completely and smoothly integrated into the throat inside wall to meet the butterfly edge. Any depression here would open a bypass path around the butterfly allowing uncontrolled inflow of air. Both these effects would be worst when the the butterfly is closed, i.e. at engine idle and as neither is controlled by the mixture screw, at the worst they could make idle speed uncontrollable.

For this reason most SU carburettors were made either with no bushes at all, or with bronze bushes that were cast into the body but actually stopped short of entering the carb Venturi. Both these designs give a smooth Venturi wall to meet with the butterfly and a tight fit round the shaft, the actual bore size being reamed to fit closely around the spindle... All of which is good when the carb is new. However, since the cast-in bushes don't penetrate all the way through, they can't be removed with a drift from either side. Consequently, if they should wear over time, then they have to be cut out. If the carb has no bushes at all (as in the HS), then the actual metal of the carb needs to be removed, allowing a bush to be inserted to mimic the cast-in type. Both these tasks are usually performed by plunging with a pilotted reamer (3/8" or 9.5mm) that takes its line from the bush opposite the one being cut. Secondly, as the bush will stop short of the Venturi, the reamer must be stopped at exactly the right depth to remove any existing bush entirely (or sufficient metal to insert one), but without breaking through into the carburettor body. This leaves the curvature of the aluminium inside the Venturi undisturbed. Once fitted the new bushes are usually then through- reamed to ensure a close fit with the new spindle.

However, the HIF4 carburettor was designed differently; its the only SU carburettor to be made from the start with removable bushes that do penetrate through the carb wall into the Venturi body! On the credit side this means that they can be drifted out easily for replacement and no reaming or drilling is required. However, it also means that the depth at which the new ones are fitted is critical since they must meet the inside surface of the Venturi exactly flush. They can neither project and jam the butterfly, nor be recessed allowing an air leak around it. As an extra safeguard against potential leaks, these carburettors also feature a moulded recess around the exterior throttle shaft entry points that houses a rubber seal. In theory then the Hif4 bush replacement is a much simpler process- drift out the old bushes, drift in and position the new. As these are internally teflon coated they should not require internal reaming either.

Rebushing
The spindle on this carb was clearly worn with obvious wear shoulders (see previous post). Its often possible to get away with simply replacing the shaft as it tends to wear more than the bushes, but my new shaft was actually a little narrower than the old  (new shaft, 7.845 mm; old shaft, 7.887) measured in an unworn area and I thought it still seemed a little loose. I concluded that really need new bushes as well.

The old bush is visible recessed inside the spindle boring at the base of the seal housing  when the seal is removed.


Its also visible on the inside of the carb and its worth checking this to make sure you know how to position the new one. I tried 3 ways to remove it:

Firstly drifting from the inside: This is a very thin bush (0.8mm wall) and this makes finding any form of drift that would actually work to tap it out without damaging the boring- quite tricky in practice. I found  5.5mm 1/4" drive socket was the right diameter but too short to drift it out completely. If you want to take this route then its probably best to machine a purpose made shouldered drift.

Secondly I tried a blind bush remover but this  failed to grip the end of the bush and simply pulled out. I suspect the bush is just slightly too long for my puller mandrel.


Finally I used a 3/8 BSF tap to thread its way into the bush and then tapped this out using a drift inserted through the opposite bush. This worked really well except on teflon coated bushes where the cover tends to shear and strip off leaving the bush in position. However once the coating has gone the bush came out fine and provided the tap is inserted square there is no damage to the bore. I think this approach might also work on the blind cast-in bushes

Once out, the  bore into which they were fitted did look a little rough. I cleaned it up a bit by inserting a 9.5mm drill bit and rotating it gently by hand. The drill wasn't a particularly precise bit at a measured 9.3mm, but it seemed to clean the odd spots of white corrosion well. 
Drill did pass through bores, insert it blunt end first and rotate gently by hand once its across the carb. The shank then acts as a pilot while the bore is cleaned. In theory I could do the same thing with a 3/8" or 9.5mm reamer but I fear it would enlarge the hole too much.

Cleaning with a reamer would be more efficient, but although I expected this bore to be 3/8", it was clear that it was actually a little smaller and wouldn't pass a reamer without some metal being removed. I really don't want to do that because the new bushes and spindle I have received from Burlen (unlike the spindle from Moss for instance) are not oversized; quite the opposite in fact. The spindle I have received is narrower that that I have removed! As received the bushes are a generous, loose fit on the new spindle. I assume that they will be compressed during installation closing them up to provide a good fit with the throttle shaft. The new bushes are 0.3777 inch external OD uncompressed- which means I would have only 2-3 thou of "squeeze" if  I reamed the hole to 0.375 inches; is this sufficient? I worry that if I were to do anything to enlarge the bore then the bushes might not be compressed sufficiently to fit smoothly around the spindle so I decided to try fitting the first bush without enlarging the hole and reconsider my approach if that failed.

The next question then is how to fit the bushes. The advice is to tap the bush in to start it using a drift larger than the bush- a wide punch or the back of a socket, to prevent the top of the bush being belled out. Burlen sell this little bushing tool to position the bush inside the bore. You can use it upside down to start the bush and then flip it over to position it. The stepped end is a loose slip-fit inside the bush while the shoulder is so-positioned as to bottom out on the carb body when the bush is in the right position. 
Burlen bushing tool and bush
Bushing tool inserted into bush- Note good match of pilot section and bush length

The problem is that this tool is designed for carbs where the bush boring does not go right through to the inside, and so is designed to position the bush short of the inside Venturi wall. How then can it be used in a Hif4? I did ask Burlen for their advice on how to do this. Although there is (as always) room for confusion I did specify the Hif4 and pointed out that its bush does penetrate the carb wall. The method I was given was exactly the same as that for HS2. In other words the bush should not be removed, but reamed out using a 9.5mm reamer- and that reaming should not go through to the inside of the carb. Obviously this would leave the stump of the old bush in position  on the inside wall of the carb. Presumably the assumption is that this bush will be correctly positioned. The new bush is then pushed up against the old using the tool above, which presumably positions it against the stump of the old one without moving it. Any wear in the old bush remnant is not a problem since the spindle will ride in the new bush. As an approach it seems more complicated. Further it involves using an expensive piloted reamer and negates the design of the Hif4 which has done away with that requirement. 

Anyway that advice came a little too late for me as the bushes were already out (as the actress said to the Bishop!). Consequently, using this tool in my carbs would leave the bush recessed app. 4mm  from the inside Venturi wall, creating an obvious air leak around the butterfly. To solve this I used a section of 12mm round brass rod to make a modified piloted drift with 2 steps. The pilot guide is narrower than the intended spindle as anything larger would prevent the bush from closing up during installation- potentially damaging the bush and/or cracking the carb body.  The pilot section is followed by a shoulder  up to the OD of the bush and this is designed to fit inside the spindle bore of the carb body allowing the drift to push the bush deeper into the bore.  The final step locates the drift on the outside of the carb and I tried to make it the correct length to place the inside of the bush flush with the inner Venturi wall. I made it by tailoring it to the carb by trial fitting and it may well be different for others, but for reference my sizes are below.

Brass bushing drift with 2 steps

Bushing as located onto home-made drift, the shoulder below it will allow it to be driven deeper into the bore and the wider section will position it at the inner Venturi wall.
Note diameters are not exactly critical but lengths are more important
Section                     Length                             Diameter
Pilot                         12.5                                 7.36
Step 1                       2.55                                 9.34
Step 2 (rod)              50                                    11.98



As a final insurance (plan B) I  selected a Draper socket which was exactly the right size to make a smooth sliding fit inside the carb throat, fitting smoothly against the wall and covering the throttle spindle holes. Putting this as a plug inside the carb should allow me to tap in the new bush and prevent it from going too far and ensuring it fits flush with the inner wall. It may even impose a rounded profile on the end of the bush for a better fit with the butterfly.

Draper 28mm socket


Socket fits pretty well inside the carb throat.
I used some Loctite on the bush to make sure its airtight. Its not going to take much longitudinal stress and I might want to remove it if this doesn't go well so medium strength lock and seal no 243 should be fine (I hope!).
I used Loctite fix'n seal on the bush
I started the bush with a wide drift to avoid spreading its top. This tapped the bush down to the level of the seal housing when I...

... switched to the stepped brass drift to push the bush in deeper up to the the shoulder of the drift. 



Any fine adjustments could be made from either end using the 5.5mm socket


The bush was inserted relatively painlessly and was clearly in about the right position. In fact its hard to know if you have put it in far enough until its in too far! Test fit with the (old) butterfly. If the bush is standing too proud it will obstruct the butterfly, but if this is the problem then you can either tap it back using the socket from the inside as before, or probably better, you can carefully file it down using a mini-drum sander in a Dremmel. Take care though because its the bush you need to sand not the carb body. In fact this sanding is really simple because the bush is so thin walled, it needs only a gentle touch to shape the bush. Taking into account the fact that the end of the bush is square-edged it might even be an improvement to slightly over-insert the bush and then carefully sand it down to the carb throat profile.


Carb bush installed.
I did find a small amount of burring on the bush top from my drifting, but this was easily removed and the spindle then slipped in easily and turned smoothly with far less play than I had observed previously. I left it overnight for the Loctite to set and then fitted the butterfly and rebuilt the rest of the carb. See next post!




Saturday, 6 May 2017

SU Hif4 carburettor removal and carburettor strip

There are a few signs that my carbs really need some attention:

Firstly, and most obviously, the idle is really rough- its hesitant and lumpy and sounds like a car that's always running  on the choke. The second problem (possibly of my own making) is that the idle speed will rise spontaneously from around 900 rpm or so to 2000, it will stick there and gradually come down. Blipping the throttle doesn't help and the throttle cam is right down against its stop. Idle will also rise spontaneously with the cables disconnected! Obviously this means that the car doesn't slow down as well as it should when you take your foot off the throttle. The lumpy running was evident from the start, I had expected a carb balance to sort that out (see my first post) and although they found a stuck piston, it didn't really improve matters. I tried to "help" by changing the air filters, using a Colortune to adjust the mixture and an air flow meter to balance the flows. I followed the various guides on the internet, Haynes and the MGB workshop manual.  I was able to improve both the idle smoothness  (although its still bad) and combustion gas colour, but it could be that after this the idle speed problem either appeared or at the least worsened.

Now researching this pretty soon showed me that this seems to be a very common problem on Hif4s- many of the reports eventually conclude that its a stuck cable or cam, or even dashpot oil level- but if the problem occurs even when the cable is disconnected and the dashpot refilled then these are not likely to be the explanation in my case. I suspect most would eliminate these possibilities before feeling the need to post and frankly I don't really believe these "resolutions"  and as the problem is intermittent, I suspect that sometimes dealing with these obvious things again does sometimes seem to have an effect but I doubt a permanent cure.  One problem I can eliminate is piston wear or poor fit. This would make the pressure chamber leaky, meaning that the piston would be held  lower for any given engine vacuum admitting less petrol and reducing engine speed; the opposite of my problem.

Anyway my considerations lead me to several possible explanations:
1. Air leak- I can't find one but many people report that warped and distorted manifolds are commonplace and they should be carefully re-flatted. WD40 spray- the usual test seems not to be particularly reliable. Anyway I can't find a leak that way.
2. Weak piston spring- permitting the  piston to rise more than it should in response to a given engine vacuum. These cars had "red" dotted springs and they can get weak- especially when hot.
3 Over-run (or poppet) valves sticking open. These allow extra air through when manifold vacuum is excessive- e.g. on overrun when decelerating from speed with the throttle closed- in other words just the conditions under which my idle tends to stick high. Again these springs can become weak and then the valves can  get gummed up or stick open and again heat can be a factor.
4. A faulty bimetallic strip inside the carb. This compensates for fuel temperature, admitting more as fuel temp rises. Its not clear to me that this can wear out, but if it goes wrong- or has been bent or installed badly maybe it could speed up idle. My suspicion is that such a problem could certainly increase idle speed; but it would stick there and wouldn't drift down again. Nonetheless this does seem to be a commonly suspected item- but I've never found a report where changing it had sorted the issue.
5. A faulty heat-shield could exacerbate all of these problems. My car has the original nasty composite fibreboard shield (asbestos???). I'd like to get rid of it although it means deviating from my goal of  "standard" condition.
6. Enrichment/choke device leaks. This device has a gasket, an O ring, a seal and a dustcap. All these older materials have been reported to fail in contact with modern ethanol fuels and leakage here would cause the rough "choke-y" idle I observed as well as  tending to enrich the mixture leading to the sooty plugs and exhaust I had also seen.

Problem 6 seems to be the most common fault - and also the cheapest to fix. You need part numbers seal AUD3577; O ring AUD3602 and gasket AUD3605. I have heard that changing the gasket for a rubber washer can also help. Perhaps if I had more time I would try fixing the choke on its own first, but as I don't wish to do this strip twice (and have to buy yet more new gaskets each time) I think I will change everything in one fell swoop as well as giving the carb and its components a very good clean. After all I may have a combination of several of these problems so a major carb rebuild does seem justified. I ordered the major rebuild kit from Burlen, this comes with new needles, jets and seals/gaskets (including the choke device seals above). It also includes new throttle plates (and spindle bearings plus seals) with or without new poppet valves (as you select) so it does address most of the possible problems. It doesn't however include the bimetallic strips or piston springs so I ordered these separately- springs cheap-ish but bimetallics seem a bit expensive! I will also check all manifold junctions for distortion and re-flat them where necessary.

As I think I have mentioned earlier, I can't leave a dead car on my drive (neighbour problems), and there is no way I can strip, rebuild, refit and retune the carbs in a day, so I'm going to get a scrap pair from AndyJennings MGB parts. He supplied the hubs and stub axles I used for my kingpin rebuild and it was a good service so I tried again and got two carbs, linkages and manifold for £100. These should allow me to strip check and rebuild the pair before doing a simple swap onto the car which should be do-able in a day. The only problem is that this used stuff isn't usually very clean...

Removing The Carbs
I received a pair of carbs plus linkages still attached to their manifold and old heat-shield. I took a few general views to see how things were attached before I started to dismantle.

Carbs front view. Steady plate (top) not apparently connected to anything- must find out where that is meant to go.

They were pretty dirty. Note vacuum takeoff for the distributor advance mechanism is on the manifold as it was with HS4 cars. I understand that this is sometimes moved to the carb itself although that's not the case with my car and these carbs match my own setup in that respect. 

Top view of dashpots and steady-plate. Vac takeof


They are still attached to the original heat-shield, made of some horrible looking composite board. This is clearly looking very sad, both in this case and in my car too. Although I do want to keep things original, I am concerned about poor heat shielding and it may be that I have to bite the bullet and fit one of the later replacement pattern metal heat shields.

Side views...


I took some general view of the linkages and springs to help with refitting

These linkages are all 2BA nuts and bolts. Throttle at rear and closest to the block, fast idle at front.


 Detail of fast idle actuation plate and the choke cable connection nipple (bottom centre) is a 5BA
The pistons aren't usually replaced in a rebuild; they come as a matched pair with their housing and they're very expensive. Wear here probably warrants a new carb. I checked that both pistons rose and fell easily with finger pressure and using the lifting lever. I will perform a timed drop check as one of the first things I do since I don't want to waste my time reconditioning carbs that are beyond help.
Pistons rose and fell easily.
Each carb is held on by 2 x 1/2" nuts, each with a spring washer, attaching to studs on the manifold. One of mine was missing its washer. The upper nuts can't be removed unless the carb is eased forwards so you need to slacken the bottom nuts as well, pull the pair of carbs forward and then remove the top nuts.
Upper carb nuts- cant be removed unless carb eased back

... and on the other carb

Nut and spring washer- 4 in total.
 The carbs then pull forwards off the heat shield/manifold and the two linkages separate as they do so. I had expected these to be fixed but in my case at least they just slid apart.
linkages- throttle top and cold start below

Choke linkage separated.
This leaves the heat insulating spacers on the heat-shield. The shield has clearly seen better days...

...and gives you two separate carbs ready for cleaning.
 I could now see the throttle flaps and both had poppet valves as expected for a carb similar to my own.
Old gasket stuck on flange- poppet valves in throttle plate.

Stripping the carbs
Unlike my experience with Dellortos, the SU carb isn't explained in a nice glossy up-to-date book. Where is Mr Hammill when you need him? Instead I found the Brooklands Guide "SU carburetters tuning tips and techniques" (Shouldn't that be "carburettors"?). This is an old book, its the best I could find but its turgid- with text that seems to have been written by Mr Cholmondley-Warner! There are no photographs (not even b/w!) only line drawings and they are very small. I will therefore try and put some more detail on the stripping and rebuild process as I experience it, but the usual provisos apply and it will be best to read in conjunction with a guide like this as I describe my experience. The text (well relevant bits anyway) from this book is available in the technical section of the Burlen fuels website. The first step is to mark dashpot cover and float chamber cover with punch marks for alignment when reassembling - I used single punch marks on carb 1 and 2 on carb 2 (duhhhh).



Marking dash pot cover and carb body no 1
The next step is to unscrew the three  screws that hold the dash pot cover (I found these really tight). I had also expected washers underneath them so I will check that (later found they are not needed here). Then...
 ...lift the cover vertically straight up revealing the spring...
 ...and any muck inside the cover.
 Remove the spring and lift out the piston.
 The needle is held in the piston by this retaining screw, before undoing it note the orientation of the needle holder in the piston as this governs in which direction the needle will be offset.

The etch mark is a small "V". In my case it points between the two air bleed holes.

Etch mark on needle holder- small "V"Note shoulder on needle is flush with piston surface.
 Remove the retaining screw, note it has a domed end and...
...pull the needle, its holder and the spring out of the piston. I will fit a new needle but reuse the spring and holder.
Next move to the bottom of the carb and unscrew the 4 screws that hold the float chamber base in place.
 Note that these screws do have washers.
 Lift off the float chamber base- might need a tap with a soft mallet. Note sealing ring in the base and this should always be changed if the base is detached. This reveals the float, jet and the jet adjusting bracket
 Some muck here but much less that I was expecting.
 To remove the jet first remove the jet bracket adjusting screw. This is the screw with the prominent cross recessed on the side of the carb.
 It unscrews and pulls out although its a bit fiddly to grab and pull. Note that it has  a shaped (reduced width) end and carries a sealing washer.


The jet adjusting bracket- bimetallic strip can then be removed by undoing the retaining screw from above.

... and remove the screw

The retaining screw has a reduced diameter end and carries a spring. It was screwed in tightly.

The jet then lifts out bringing the adjusting bracket with it and then...



the bracket is removed from the jet by swivelling the holder to align the lobes with the cut-outs in the bracket.


The two separate
Returning to the carb float bowl the float is held in  by a spindle screw
and the head of this screw is visible outside the carb
Undo the screw and remove it- it has an aluminium sealing washer just below the head
Loosen and remove the washer- it will be replaced.
The float then lifts out. I shook it to confirm that it wasn't leaky- no fluids inside! The needle valve is below and the needle simply lifts out. This was an all metal needle and will be replaced with a viton tipped modern equivalent from the rebuild kit.
The needle valve seat then unscrews using a 2BA socket

The seat lifts out and will also be replaced.

The main jet is retained by a central nut- 5/16 WW and easily undone with a socket
Poor focus here- sorry, but the jet bearing retaining nut then lifts out...

... and the jet bearing itself then pulls out - note the aluminium sealing washer on the jet bearing body. Later carbs have a jet bearing that doesn't have this washer so if its missing its not necessarily wrong! The test is to fit the jet bearing without the washer and check it doesnt protrude above the bridge under the throttle slide. If it does then it will need a washer to lower it. My newer carbs as fitted to my car are type FZ1229 (f and r) and these dont have the washer.

The next step involves releasing the two springs that drive return for the throttle and fast idle control levers/plates. I took some pics to show how the two were fitted. The fast idle spring is held between the operating cam plate and the fast idle spindle cover. Its top is hooked behind the cam plate and its lower end behind the cover plate. This spring is removed first but its bracket also retains one end of the throttle return spring. The latter is visible in the pic below at the top- the lower end of the fast idle spring is hooked into the fast idle cover plate. Note that the throttle spring is also held by the cover plate. 

Fast idle spring- note upper end (hooked) located behind cam plate centre right. Lower end (straight) located behind lug on cover plate. centre left. The throttle spring extends its straight end to locate behind the cover plate centre top

Upper end of fast idle spring (picture inverted)
The fast idle cam is held by a column nut, removed with a 3/16 WW socket.

The nut then lifts off revealing the tabbed lock washer beneath.
Remove the lock washer and ease the cam plate off the spindle, allowing the spring to expand as it comes off.
Remove the spring and the dished seal-covering washer beneath

The locations of the throttle return spring ends are clearer now that the fast idle control system has been removed. Note how its ends locate onto the fast idle cover plate and the throttle lever.

Side view... throttle return spring located in throttle pillar
The fast idle/choke system can be further removed- the small seal lifts off the spindle...

Remove the spring and...

... the two screws holding on the retaining plate/cover can be unscrewed.
 No washers here.

Cover removed
 and the fast idle system can be pulled out- note that the whole thing comes out not just the threaded section. I found it was pretty tight but it came out eventually.
 The inner shaft can then be pulled out of the outer- note the dust shield that was over the threaded end and the sealing ring on the outer shaft. There is also a gasket but this remained fixed to the carb body. These components seem a very common cause of the unstable idle speed I have been experiencing so should always be renewed.


Returning to the carb body the column nut holding the throttle levers undoes with a 3/16 WW socket.


There is a thick washer under the nut. I was expecting a tab washer so I'll check before reassembling. The  throttle levers can then be eased off the spindle. I found they were very tight.
Lever is in 2 parts... part 1 and....
... and part 2. Note sealing washer around spindle. In this case it was quite brittle and cracked.
The next step is to remove the throttle butterfly. There is a mysterious insistence on marking it and the "flange" although I don't know why. I'm not refitting this butterfly anyway so pictures should help more in reassembly. The butterfly screws are split and the ends bent to prevent them loosening in service.
Squeeze the ends to restore them and unscrew.

Once removed, slip the butterfly horizontal and slip it out avoiding the poppet valve
... and withdraw the butterfly

This leaves the spindle which...
... just pulls out of the body. Again pictures should help reassembly.
One seal came off with the spindle. The other stayed on the carb but both were in poor condition.
The spindle has obvious wear shoulders where it passes through the bushes.

The seal was brittle and broken.


Removing the seal from the other side (it was in similar condition- the edge of the bush is visible inside the throttle shaft boring (and also inside the Venturi).




That just about completes the strip. I now need to give everything a good clean and assemble my new parts. The spindle on this carb was clearly worn. Its possible I might have got away with simply replacing the shaft but the new shaft was actually a little narrower than the old and wear was still evident when it was slipped in. It seems therefore that I really need new bushes as well. The bushes are visible inside the spindle boring when the seal is removed. I will cover rebushing and reassembly in further posts.