MG-Cars.net

Welcome to our resource for MG Car Information.

Recommendations

Parts

MG parts spares and accessories are available for MG T Series (TA, MG TB, MG TC, MG TD, MG TF), Magnette, MGA, Twin cam, MGB, MGBGT, MGC, MGC GT, MG Midget, Sprite and other MG models from British car spares company LBCarCo.

MG Midget and Sprite Technical - Problems fitting high torque starter motor

Hi all!

Recently bought a high torque starter motor (pre-engaged type) for my 1430 midget from Cambridge Motorsports.
very pleasant company to do business with and the starter arrived in the Netherlands in no time. saturday I planned to fit it, but I encountered some serious problems
Fitting was quite easy, just need some triple jointed fingers to get to the lower mounting and I had to remove the dizzy. After fitting I started the car and heard only *Click* from the solenoid...checked the wiring which is pretty straightforward (I left the solenoid on the bulkhead in place, so I didnt remove the link wire on the starter, as per instructions). Measured power on the starter when starting, all OK. Checked everything else, not that there is much to check, link wire was securely in place.

Removed the starter from the car again and tested it with some jump cables, and it turned OK! Refitted it to the car, tightening all the bolts semi-tight this time to save time. Started and the engine was spinning this time. Happy it ran, but with a strange feeling, because I changed nothing!

Tightened everything up, refit the distributorstarted...*Click* and that was it again...Grrrrrrr

I have reverted to the old starter again because I need the car.

A day wasted...well, almost wasted, because I fitted the electronic ignition which is working really well!

Any of you had (or heard of) similar problems???

Cheers,

Eddy

Eddy Veuger

Eddy - it sounds like the solenoid is working ok (that's the click)..

When you refit the dizzy is it interfering with the big fat starter cable? If its like mine (and the photo looks exactly the same but mine comes from Power-Lite) then its just a single connection (power) to the stud at the back...

You know that it spins on the bench, you know that without the dizzy it spins the engine so logically it must be something either through tightening up the 2 bolts that hold it in place or its been disturbed when you refit the dizzy.

Whats the chances that you have a break in the starter motor fat wire that you cant see?
Toby Anscombe

Eddy,
Did you accurately compare the length of the bendix gear shaft on the old and new starters. It sounds like the new one works when out of the car, or possibly even when loosely installed. But perhaps when fully tightened up it is jamming so you just hear the solenoid click. Some starters use extra spacers between the starter flange and the engine backplate to correct this. Also check that the bendix gears are the same diameter and number of teeth.
Guy Weller

Eddy can we get a picture?

A pre-engaged starter ought to have its own connections and I cannot see why it needs the original solenoid left in

and there could be a problem with teeth numbers as Guy says

there is a sequence of events with a pre-engaged, first power from the ignition switch comes down the small connection energises its solenoid and moves the drive gear into mesh with the starter ring.
THEN the solenoid generally switches on the heavy duty contacts inside that take power through the heavy duty wire from the battery and starts the motor turning, which turns the flywheel.

I would not leave the original solenoid in the sequence to "get in the way" of the intended power run, from battery to earth.

So a picture would help us to see what causes your problem
Bill

OK,

I'm at work now, so not in the position to make a picture, but I will do so tonight.

In the fitting instructions are two options, leave the original solenoid in place or use the one that's build in. By removing or attaching a link wire you can choose between these options.
The fact that it worked once on the engine (loosely connected) indicates that the electrics are in fact working. But I agree that it would be better to lose the original solenoid, if only to eliminate it as a point of failure and I am planning to rewire.

How many teeth should the gear have?

It can't have anything to do with the dizzy, because I tried with dizzy in and with dizzy out...same result. I also tried to slacken the bolts again, but no luck...

Anyway, I got a reply form Cambridge asking me to do some measurements, so I'm guessing the old starter motor will come out again tonight....I'm getting quite good at it!

Cheers,

Eddy
Eddy Veuger

So back to your original post:

loosely fitted the gear turns under power
tightly fitted nothing happens except a click from the solenoid

so either:

The motor is faulty and when bolted in tightly refuses to turn but when the locating flange is loose some distortion may be released

or

the drive gear is forced up against the flywheel and may need an adapter ring for the application in an A series engine

or

they sent you the wrong part

measure the distance from the face of the engine backplate to the flywheel teeth and check that against the flange face of the starter to the gear teeth on the drive

to operate the motor the solenoid needs to "push" the drive forwards and "make" the motor begin turning. If it cant push it cant turn either.
Bill

I worked with Cambridge Motorsports to produce the Spridget pre-engaged starter so unless they've changed supplier or send you the wrong one it should work fine. I doubt they've send you the wrong one as it wouldn't fit. It also should not need the packing piece.

Toby is right, the click is the starter motor solenoid flicking the gear out + what Bill says

I forget whether or not I retained the solenoid on the car.

One thing to watch is that the starer motor isn't contacting the bodywork such that it's preventing it being bolted in correctly.

I'd bench test (without the bench) by sticking it on the floor and using jump leads to make it work.

I agree with Toby that possibly the live lead is suspect, having broken down internally.

I don't have an electronic image of what a Cambs starter should look like but if I can find the original photo I'll scan and e-mail it.

Daniel Thirteen-Twelve

So, just removed the old starter again. Measured the engine block surface to ring gear, which is 28,5 mm.
Compared this to the starter and it looks like that's OK. Gear on the starter not touching the ring gear when disengaged and it comes out far enough to positively connect the gears.

Test with jump leads I did already yesterday.
It works OK outside of the car.
Fact that the old starter works just fine rules out any electric issues. Lead is no solid-core, so a breakage is out of the question.

However...the old starter has a 10 teeth sprocket and the new one only 9...

This would explain the erratic behaviour, the one time it worked the gears where just perfectly in line, the other times they just hit each other but didn't engage.

What do you guys think?

Cheers,
Eddy
Eddy Veuger

On my Hi torque starter it has a built in siloniod, and I had to make a jumper wire that looped back on the starter....did you have the same thing...Here is the on-line instructions I used

http://ferraris.server273.com/solenoid.html


1480 is alot of motor...granted it does start with the old lucas starter I guess,

How big a cable are you using,

Im guessing the starter is sitting flush with the rear engine plate...not caddy wonk


prop
Prop***The End in 2012

Hi Prop,

Same here, there is a jump wire to the solenoid, works OK. Still I'm planning a rewire, just to take out the old solenoid out of the equation.

Cable is 40mm2. The new starter pulls less current than the old, so no worries there.

Cheers,

Eddy
Eddy Veuger

Wow eddy,

That is intresting, I got to go with the crowd ...manufacture defect.

what bridge are we all meeting at, I cant jump till after 7pm...LOL

Prop
Prop***The End in 2012

Just re-thinking, you never had problems with your old starter doing the "click" routine, if not Id say the selinoid is okay, esp. if the old starter was pulling more juice then the new one.

prop
Prop***The End in 2012

Can it be that you have a non original flywheel which causes the new starter not to line up with the teeth?
(just try to think of every posibility..)
Arie de Best

I thought about that too Arie, but then the old starter wouldn't work either wouldn't it?

I'll await the verdict from Cambridge Motorsport about the nr of teeth on the sprocket...

Cheers,

Eddy
Eddy Veuger

Shouldn't the standard starter motor have 9 teeth?
Daniel Thirteen-Twelve

I have no idea! There are so many mods on this engine that this could as well be one of them! All I know is that the old had 10 teeth and the new just 9...
Eddy Veuger

I thought that the old starter did work OK - it was just that the engine was reluctant to start so it had to continue churning the engine until there was no energy left in the battery.
So if the old starter had a 10 tooth bendix to match it to the flywheel, maybe Nick did have a non-standard flywheel as Arie suggests.

Can you perhaps swap the bendix gear from the old starter to the new one?
Guy Weller

Just checked my spare starter for my 1275 it has 10 teeth also.

Moss book confuses further by mentioning under the spare parts 9 and 10 teeth - but fails to explain where required.

New game - Whats yours got ?

Richard.
richard boobier

Guy pre-engaged has no bendix

If the gear engages sometimes (mostly off engine) I would guess the ring gear may need reversing so the lead in angle is facing the drive gear.

The lead-in angle ought to be on the inside of the bellhousing towards the gearbox for a bendix gear driven start but for a pre-engaged it would need to face the hole the starter bolts into

In the present configuration it will hit a flat face without a lead in unless it is very lucky and hits the space between teeth
Isnt this an issue when deciding which starter to use for K series engined Spridgets, one type works from the gearbox side and the other works from the engine side
Bill

Bill, I didn't know the ring gear also has a lead in angle. Thought it was just the gear on the starter. I have to check that in the morning.
But surely Cambridge Motorsports would know that and it would be in the instructions as well as it describes fitting the pre-engaged type where a inertia with bendix used to sit, or am I being naïve here...

I'll check that in the morning!
Eddy Veuger

My factory parts manual only lists the 9 tooth pinion for both 1275 and 1500 engines.

The pre-engaged starter motor engages the ring gear from the opposite side to the bendix motor. Since the bendix gear is spinning as it contacts the ring gear it does need that lead in. However the pre-engaged motor engages the gear and then spins so it does not need a lead in and will engage a flat faced ring gear.

With my own car I went straight from the bendix to the pre-engaged on the same flywheel without any problems whatsoever.
Daniel Thirteen-Twelve

I seem to remember many if not all starter gear teeth having a lead in

the picture shows a part of the page in my Honda Accord Haynes manual (p145 if anyone has one to look at)

It clearly (in real life!) shows a lead in angle on each tooth, not awfully clear in this but you can see the angle on the teeth

I am not saying this is all that can be causing the problem, but I think it worth a consider...


Bill

Yes, both the Bendix and pre-engaged have a lead in on the gear. It's only the bendix that also needs a lead in on the ring gear as well.
Daniel Thirteen-Twelve

Not ONLY the bendix

the Honda certainly does have a lead in
Bill

Its sure looking more and more like your going to need a 10 gear, If so I cant imagine that being a real issue with the company that makes them, most likely you will have to send your old starter in so they can match up the gear profile


Before you do that look to see if there is an est. starter/alt rebuilder in the area, Im certian they could modify it to fit and mesh correctly...

Sometimes 9 teeth are not enough, to take the home comming queen to prom....


Prop
Prop***The End in 2012

I know that I raised this question of the gear being the same diameter and number of teeth. Clearly it needs to be designed to mesh properly if it is going to work. But I am not so sure that the actual number of teeth, 9 or 10, would make that much difference. It would affect the gearing of the starter, but so long as the gear diameter is right, then I think the "extra" tooth would not itself prevent it engaging. The reason I say this is that the teeth are not exactly a close fit to the ring gear and I think there is enough design tolerance to allow for either to mesh. What is more important is the diameter of the gear and the bevel(s) to allow them to engage

Bill, I hadn't realised that the new one was not a Bendix. My suggestion of swapping the gears over clearly wouldn't work!
Guy Weller

Eddy,

Are you certain the link wire is Ok and that when the starter receives power the gear is flicking out ok?
Daniel Thirteen-Twelve

Hey Eddy,

You owe Me...LOL, I took some photos of my stock 1275 fly with the Hi torque ingauged on the teeth

the starter has 9 teeth...as close a mash as I was able to see, Id say your problem is a diffreant fly wheel ring, i hope its not an engine out situation, my type of luck it would be...

2 photos....back to work, hope this helps

Prop


Prop***The End in 2012

and the secound

prop


Prop***The End in 2012

Eddy

I had (have) a similar problem on my racer. Try refitting the hi torque starter again - if it's the same, try rocking the car gently in top gear then try the starter. Mine seems to be getting slightly better -but it's early days and currently the engine/starter are out of the car. Best of luck.

Steve
Steve Collinson

Lucas changed from 9-tooth to 10-tooth at some point (in the late 80's I think, before the change to pre-engaged ones), but the flywheel and ring gear remained the same. Don't ask how that works, but when I needed to replace the starter on our 1380 off came the 9-tooth and on went a new 10-tooth. As Guy says, there is quite a bit of clearance, so maybe they just used some if it! From days of calculating gear tooth meshing clearances to within a few thou, it seems a bit strange, but??

I queried it at the time of buying, and was told that it would work fine, which it did and does. Apart from turning the engine over a bit faster (10/9 = 11% quicker, which is very noticeable), it's been working very well for 6 years.

I have never understood how a front engaging starter will work reliably with a ring gear that is designed for rear engaging - there are no leading chamfers on the engine facing side of the ring. Maybe it use the same magic as a 10-tooth pinion - all a bit bizzare!

What is the recommendation about the ring gear? Just turning it round does not put the chamfers in the right orientation for a pre-engaged starter that will be turning in the opposite direction to the rear-mounted inertia start type??

I think I will stick to the standard Lucas version ..... ducks and leaves the room very quickly!!
Richard Wale

Sorry, now having read my reply, I missed out that I think problem is that it could be the starter is not able to go into mesh, because there are no leading chamfers on the egde of the ring facing the starter pinion.

Richard
Richard Wale

Richard,

Daniel pointed out earlier in the thread that pre-engaged ring gears don't have lead in chamfer and that is my experience with engine fitted with pre-engaged as standard. Regarding swapping the ring gear round, that would not be an option on the A series rings gears I've fitted due to the step in the flywheel and ring gear.
David Billington

Quote Richard W - 'I have never understood how a front engaging starter will work reliably with a ring gear that is designed for rear engaging - there are no leading chamfers on the engine facing side of the ring.'

My Frog came with a hi-torque front engaging starter. Two months into ownership it failed with similar symptoms to Eddy's - it either just clicked or spun without turning the engine over.

On removal the pinion was found to have large chunks of its teeth smashed off, presumably due to the lack of a chamfer on the flywheel teeth to guide it into mesh cleanly.

I binned it and went back to a standard Lucas unit. It has no problems starting my 1293 with a CR of 10.1:1.
Jordan Gibson

Jordan,

The other possibility is the starter was driving the pinion too early and overloading the teeth causing them to break. Normal operation would be for the pinion to be pushed into engagement with the ring gear by the solenoid then the power applied to the drive motor. Early application of the drive motor power would mean the pinion is not fully engaged, that could either be the fault of the starter or the design of the conversion.
David Billington

as I said

there IS a chamfer on the flywheel on the Honda

it uses a pre-engaged starter

with a chamfer (lead-in)

and in my opinion Honda, whose engineering capabilities are second to none in the motor industry (in my opinion!) do not dismiss the wisdom of paying the required extra dosh to have the starter operate easily by machining a flywheel blank

if you slam metal against metal there has to be a result eventually

if you make room for it to slide past the opposing metal the result is far likelier to be a happy one

Bill

Bill,

I'll see your Honda flywheel with chamfer and raise you 1 x Rover 1.4 K flywheel, 1 x Caterham 1.6 K flywheel, and 1 x Lancia Thema turbo flywheel all without chamfer.

Granted Honda have a good reputation for engineering, maybe the devil is in the pinion detail as the 3 ring gears above all look fine with no sign of mangling so the pinion must slot in sweet prior to rotation.
David Billington

My personal opinion would be not to bugge* about with this hi torque thingy anyway, but we were just trying to resolve Eddy's singular problem caused by actually fitting one.

p'raps we need to see what the makers come back to Edy with

I still wonder if the fault is in the casing and the relationship of it with the armature when bolted in tight

Problem is that without pictures we're all only guessing

I still like the Honda engineering approach :-)
Bill

Having re-read the OP I gather the starter, when fully fitted with dizzy in place, didn't work, but worked out of the car when tested with jump leads. On refitting the motor worked, but when the dizzy was refitted it didn't. I can't imagine a short in the area would go unnoticed but something odd is happening.

Eddy mentioned that the original solenoid is still used so the new starter is wired so its solenoid is energised when the main lead from the original solenoid is energised, no problem there, but I would want to diagnose whether the new, pre-engaged, starters' solenoid was activating when the "click" was heard. As there are 2 solenoids in operation both must operate to turn the engine, the original must function to operate the 2nd on the new starter. In the case where "click" is heard it would be interesting to know whether the new starter solenoid can be felt to operate, as the first must to generate at least one "click". Whether the time to energise the 2 is long enough to hear 2 discernable "clicks" I don't know.

Bill,

Regarding Honda, you might find the following amusing. It was sent to me a few years ago by a mate in the auto industry that works occasionally with Ford and he said it was too true.


Ford Potatoes

How a Honda employee bakes a potato:

* Preheat new, high-quality oven to 350 F
* Insert a 1.0 lb Idaho potato
* Go do something productive for 45 minutes
* Check for doneness, then remove perfectly baked potato from oven and serve


How a Ford employee bakes a potato:

* Conduct market test with suppliers in Mexico, Brazil, and Turkistan to
supply 0.75 lb potatoes, choose lowest cost supplier with best Wings tix
* Change to incumbent supplier of Idaho potatoes, insist they meet Turkistan
pricing with 3% annual price reductions
* Upgrade to 1.0 lb potato, insist supplier erred by pricing for 0.75 lbs as
instructed when he knows Honda uses 1.0 lb potatoes
* Instruct potato supplier to preheat the oven to 350 F
* Demand that the supplier show you how he turned the dial to reach 350 F,
and have him come up with documentation from the oven manufacturer proving
that it was calibrated properly
* Review documentation, then have supplier check the temperature using a
sophisticated temperature probe
* Direct supplier to insert potato and set timer for 45 minutes
* Have supplier open oven to prove potato has been installed correctly, and
request a free study proving that 45 minutes is the ideal time to bake a
potato of this size and variability due to
orientation within the oven
* Request a Six Sigma Study showing variable cook times for various potato
sizes and orientations
* Check potato for doneness after 10 minutes
* Check potato for doneness after 11 minutes
* Check potato for doneness after 12 minutes
* Become impatient with supplier (why is this simple potato taking so long
to bake?). Demand status reports every five minutes.
* Conduct Value Engineering session and new market test
* Change to 0.9 lb potato because customers will only notice if potato
weight is reduce to 0.85 lb
* Check potato for doneness after 15 minutes...
* After 35 minutes, conclude that potato is nearing completion. Pass through
Gate review reporting all Green status.
* Congratulate supplier, then update your boss on all the great work you've
done, despite having to work with such an uncooperative supplier
* Remove potato from oven after 40 minutes of baking, as a cost save without
loss of function or quality versus the original 45 minute baking time.
* Serve potato
* Wonder aloud what on earth those Japanese folks are doing over there to
make such good, low-cost baked potatoes that people seem to like better than
Ford potatoes
David Billington

LOL

so true David, so I was told when training on Ford products back then...

I raised doubts about the use of two solenoids (redundant failure path built in)

as you know the solenoid on the unit is enough to work it, safely and efficiently, putting two in the line just makes a fail more likely

but even now I can remember a couple of starter that tested OK off the car and failed in situ

always had the warranty honoured though, but it would be harder for Eddy to get the replacement supplied, over there in The Netherlands. Which was why I have tried to come up with as many checkable things as we can before he has to.

Something that has occured to me though.

As you know the solenoid pushes the gear into mesh (whether with or without a lead-in) and at the end of its travel then operates the motor switch. I wonder if when it is fitted the gear isnt being pushed entirely far enough to engage the heavy duty motor switch for some reason. Which might necessitate the use of a spacer to allow enough forward motion...

millimetres might mean enough!
Bill

OK, I'm back! Been on a short business trip, but yesterday eve I spent some quality time under the Midget again...
I took off 1,5 mm of the mounting plate and slotted the top hole of the starter so that the gear moved away from the ring gear. Camebridge motorsports thought this would do the trick.
But no luck...still hits the ring gear without engaging it.
As I planend to fit a 5 speed in a couple of months I have refitted the old starter and will wait until I have the engine out, then I can see what's happening exactly, because this is driving me mad! Had the bugger in and out at least 8 times now and I'm fed up!

So story will be continued in spring!

Cheers and thanks for all the help!

Eddy
Eddy Veuger

This thread was discussed between 19/01/2009 and 29/01/2009

MG Midget and Sprite Technical index

This thread is from the archive. The Live MG Midget and Sprite Technical BBS is active now.