Welcome to our resource for MG Car Information.
MG Midget and Sprite Technical - De-dion back end?
I'm part way through rebuilding my Sprite with shiny new metal and then a 1.8 VVC K series. However, my mind is turning to how to beef up the back end to handle the torque the engine will produce.
From what I can gather I have 3 options.
1: Live axle. Beef up the back axle I have. This will probably end up as being fairly expensive with an LSD, uprated shafts and so on, but will allow me to use off the shelf parts from Frontline and the like.
2: IRS. Involves a fair bit of cutting out of good metal and then making/designing a new set up that fits easily and more importantly works well.
3: De-dion. Allows me to use pretty much any diff I like and as such any ratio. Fitting some coil overs could be a faff, but no more so than with the IRS. I'd also not have to cut much, if anything out. However, I would have to make somewhere to fit the trailing arms to.
So what do you all suggest? Any hints/tips/advice/warnings welcome!
|Beef up the old axle|
-Proven to work
-no chance of a bad handeling car due to geometry c#ck ups
If you want to go diffrerent go IRS
Both other than std axle options involve a lot of work so why not do it right if you
|and watch out for Mr IVA. points number is low on a K anyway....|
go for it - get a nice cheap conversion going then we can all have one :p
|I like the idea of a De Dion axle in that it keeps some of the character of the original suspension, assuming you keep the cart springs or in my case 1/4 elliptic springs, but alleviates axle tramp and provides and potentially bomb proof diff assembly. I was looking at the idea of mounting the diff unit, such as Sierra, on a bolt in tube running from side to side, the tube being welded to plates fitting between the triangular gussets that the lever arms dampers normally bolt to. That would give the 2 damper bolts and the radius arm bolt to handle the loads and possibly the mounting plates could extend down to connect to the spring plates as well for an additional tie in. Not gotten any further with the idea though as too many other things to do.|
|Rob, what makes you say that I may fall foul of the IVA? The monocoque won't be modified and I'll have 9 of their 8 points anyway from the shell (5 points), the steering (2 points) and the rear springs (another 2 points).|
Oh, besides which, my car has never been registered in this country. I've been told by my local DVLA office who will handle the registration that they only care about the shell ID and not about that of the engine, gearbox and so on. All they've asked for is an MOT and insurance to allow them to register it.
David, I quite like that idea. I was contemplating welding plates on the back of the transmission tunnel to hold the diff in place. Means some welding on the car, but it's probably the lightest option.
|with a K series & ford type 9, you have 9 points. If you modify the axles (2 points) and suspension (2 points) with an independent rear end, then that's less than the 8 you need (using 5 for shell and 2 for steering)|
though it's not *both* axles you might be OK. I'm not saying "It will fail don't do it" - (in fact the opposite - sounds like a very interesing engineering challenge & a good idea) just making sure it'll be ok if you go down that route :)
|It sounds like you've made up your mind already.|
Something to think about is while option 1 has an LSD neither option 2 or 3 does, why is that?
Also you are not really considering like for like.
The options ought to be:
Live axle (retaining existing casing)
Those are your 3 choices. You've linked coil over's to the de-dion choice but you don't mention them with the live axle.
Probably your real issue has nothing to do with suspension but whether or not the existing crown wheel and pinion can cope with the power output you are expecting from the K-series.
LSD options for 2 and 3 should not be a problem with the likes of the Sierra diff unit as they are readily available. Probably more so than for the A series diff due to more recent manufacture, and I expect higher numbers made.
Should have said that I didn't include the LSDs for De-Dion or IRS due to them being so much more available and so much cheaper. With the live axle it's more involved, although the diff would be re-built anyway so it would be fitted then. It would almost certainly cost more though.
You're right about the coil overs, they are an option for live axle and De-Dion, but pretty much required for IRS. I think you may be right about the crown wheel and pinion part of things. I only intend building the back end of my car once and I don't want to be worrying about if it can handle what I have in mind at any point. Power wise, it needs to be reliable at anything up to 200lb-ft. I doubt I'll get near this, but if it's safe to that level I need not worry.
Is there any way that the existing axle will handle this without spending a king's ransom on it? I, as with most, don't have buckets of money to throw at this, so the cost of things will have an impact on what I do.
|midget TranX web site the Sierra diff is|
|Whilst your point about new prices is quite correct, there's no way I would bother buying a new Sierra, for example, diff/LSD as the condition and availability of second hand ones make buying anything a bit silly IMHO. Although the Sierra stuff is strong, it is quit heavy, as you say. For this reason, I'd probably look at smaller maybe MX-5 type stuff.|
With regards to the power output, what sort of torque would a turbo A series produce? I've never had a turbo'd one and only ever did NA tuning on mine in the past.
As to the weight of a non-midget set up, by the time you've got anti-tramp bars, a Watt's linkage/RTL and the associated bracketry thrown in, I don't think a well thought out De-Dion or IRS will be that much heavier. Heavier, yes, but not by too much when you consider the advantages to be had. Also, although it will weigh more, then should be less un-sprung weight, which is never a bad thing.
|ASeries turbo give approx boost x n/a torque; maybe 120 to 150 max? Bob can tell us. I saw 170 lb-ft on my 1.8 K MPi at Aldon recently.|
IIRC Russ' K showed 209 bhp but 'only' 159 lb-ft. In competition he used LSD but weakness was cwp. I helped him change diff on several occasions. It appeared that crown and pinion. became separated with applied torque and had a short milling session; I have pics.
|Anthony, that is exactly the problem I wish to avoid. I have every intention of getting my car to tracks and maybe the odd sprint if I feel brave enough, so I don't want things to start failing. Especially since this will also be my only car as I have works van for during the week.|
Sounds like my target of being able to handle 200 lb-ft is about right. Any idea what sort of torque a supercharged K might put out? Just in case the vague plan in the back of my head comes to fruition one day.....
PS, pics would be good if you can, if only so I know what to avoid!
|KT's torque... well anything from over 200 to 250 lb-ft should be quite easy; from there, it's only a matter of how much boost you can apply before the engine's had enough. My preference (if I were to go this route) would be around 1/2 bar for around? 225 lb-ft and enjoy the surge from 2K to 5K... can't imagine many cars keeping up with that.|
Tne main issue, as you know, is applying this to the road. I should have mentioned Russ was using 032Rs, which are quite sticky. You will need similar if you're not going to be spining the rears in the lower gears.
I'm using std diff and shafts at mo, and avoid WOT in 1st gear and am careful not to spin the wheels in the lower gears; not that this means the I don't have major benefits with the K...
Here's a pic of Russ' CWP... Sept '02 at Eelmore sprint. This in there looked so sharp, we didn't want to test with our fingers...
|Our K is a VVC with 185BHP (sorry, don't have the torque) and has done 9 years of fairly intensive autotesting ... usually shared by at least 3 drivers who show no mercy - clutch dumps, wheelspin (lots) etc. All while running medium Formula R's. |
It broke the first diff in the second event as we hadn't had time to fit the LSD - the spider gears seized then shattered. Fitted the LSD, since then no problems. I notice that the new CW&P's (which we fitted) have a bigger angle on the bevel than the originals. That may or may not affect the wedge effect Anthony mentions.
Whatever the reason it's held together, and we are fitting the same set-up on the K turbo project. That will try it a bit more.
|BTW - forgot to mention... Russ had a mock-up of B-banjo CWP/Diff for the Midget axle, but did not proceed as his setup seemed reliable enough.|
I have encl a pic... BTW would shim/spacer the diff to banjo housing so that the shafts are in correct plane...)
(BTW - I am interested in De Dion ideas... since it's a simple geometry to design compared with IRS and so on... would be interested to see design / pics...)
|If your keeping the standard axle/diff, you can reduce the chance of the pinion moving relative to the crownwheel(like Russ's- I saw his car do the same thing at Wiscombe!)by using a solid spacer and shims and not the collapsible washer system that most reconditionners use .|
|A similar problem I faced with the installation of the 2.8 V6 in my car. I chose to install a different rear axle housing. Much stronger and actually much less expensive in the long run than upgrading the original axle. I've looked at IRS as a possibility, but clearances are very tight under the Spridget. I can't imagine a De Dion fitting without removing the fuel tank or making it much smaller. I used a GM axle but those aren't going to be commonly available over on your side, but I'm sure you can find a suitable replacement. A couple of things to consider when chosing a replacement or IRS. Rear brake cylinder or cailper bore size. Make sure they are compatable with your existing master cylinder or you'll be adding more work to get a well balanced brake system. Also make sure that the gear ratios that might work best for you are available in the housing you choose as well as a reasonably priced LSD. |
As for the IRS, I did some initial design ideas for a mounting cage and was looking at the front hub carriers from many FWD cars as a rear hub carrier to keep costs down. Most use a ball joint as the lower mount and those that come from cars using a McPherson strut usually have two holes in the upper end that would work as mounting points for the upper control arm. The lower control arms would be tied to the same point as the existing front spring hanger. Here's a sketch of what I had in mind if it might help.
Your IRS set up is what I had in mind if I was to go IRS. As you say, space is a little limited under such a small car. The brake issue won't be a problem on my car, as I'll be fitting all new brakes and maybe going for twin master cylinders, so any brake set will be easy to control.
With the De-Dion set up, I was planning on moving the diff forward a little to gain more room to run the De-Dion tube between the diff and the tank.
The main problem I can see with the De-Dion set up is having somewhere to mount the trailing arms if I go to coil overs. And since I would like to, I'm trying to give this some thought. One could be mounted to the front spring hanger and I'm thinking of copying Caterham and sending the other link backwards as though it was a Watt's linkage with a pivot point on the hub carrier somewhere. A bit like, or pretty much exactly like the pic. I'd then probably still use another Watt's or maybe Mumford link for lateral location.
What do you lot think of it as an idea? Worth a little investigation?
|I'm not a fan of the DeDion rear suspension. A great design for it's time with bias ply tires, but with radials you can get a lot more grip if you can vary the camber during suspension movement which is a down side of both the DeDion and a live axle rear. The weight savings in unsprung weight isn't that much over a live axle anyway, just the center section which is pretty light for the original MG with the alloy gear case. In addition you can design in some compression steer into an IRS if you want which isn't as easy with the DeDion. |
Either DeDion or IRS would need some sort of cage to mount the diff, so that would incorporate the upper mountings for the coil over units.
As for the trailing arm links, those need to be parallel to each other and the same length and mounted on the same side of the axle. To mount one forward and one to the rear works on an IRS where the carrier is free to rotate a bit, but on a DeDion you don't want to impart any twist to the DeDion tube. That becomes a bit of a problem then with the limited space available under a Spridget and still maintain the longest arms possible to reduce fore and aft movement during suspension movement. That's why I favor an IRS with a lower A frame with poly bushings to allow for some deflection and an upper locating arm to control camber. Torque reaction can be controlled by a single trailing arm extending from the original spring mount point to the lower A frame just inside the ball joint. With an arm that long and using poly bushings there would be minimal fore and aft movement and the bushings would reduce any bind.
|Whilst I agree that the weight saving isn't massive over a live axle, not having a diff bouncing up and down is an advantage.|
I know that a more traditional De-Dion would mount both arms forward, but using a Watt's linkage as Caterham do will eliminate any fore-aft movement at all and the pivot means there is no twist in the De-Dion tube. It does mean you can't have any anti-dive built in though.
I do like your IRS idea though, so don't get me wrong. I'm in the process of deciding which will be the best compromise for me. Whilst IRS does have many advantages, designing a good set up is a lot harder than with a 5 linked De-Dion/live axle, which is relatively simple. Although fitting a few rose joints to things can allow a fair range of adjustment with the IRS.
The camber can be set as you want it with the DeDion by appropriate design of the bearing carriers and their mountings, in fact Caterham offer wedges to allow the camber to be changed to suit the tyre type. Regarding camber change on IRS you usually get a trade off as good camber on the outer tyre when in roll result in a poor camber on the inner tyre, DeDion has the benefit, as with a live axle, of little camber change in roll and the inner and outer wheel staying upright.
|David, you are correct, but the camber setting is fixed on both the live axle or DeDion rear so it is necessarily a compromise setting resulting in a bit less contact patch when level to gain some added traction when cornering. The IRS does usually limit the traction on the inner tire when cornering to gain additional traction on the outer tire, but then most of the force and weight of the car is being carried on the outside at that time anyway, the inner tire is almost along for the ride. As for both tires being upright in relation to the ground, that's not necessarily the best angle for modern radials unless you run a lot of air pressure which is fine for a competition car but rough on the street. |
I believe that an IRS can be designed to give maximum straight ahead traction while maximizing the traction during cornering by tailoring the camber change curve during suspension compression and rebound. I admit it's far outside my engineering abilities but there are software programs around to help the average guy design such as suspension.
|That's part of the problem I face. I'm not sure I have the skills required to design, build and set up a good IRS set up. Given time and a fair bit of patience I don't see why it won't be possible, but a live axle or De-Dion is a much simpler thing to design and build. And that is quite important since I may well be making some/all of the suspension parts myself.|
|stuff simplest - go for the best :)|
You'll always be thinking ooh i could have made it that way
I wasn't sure I could fit a K series but it turned out I could :)
if there's no rush, then IRS may be the way to go - also, it's been done before so you can see how others have done it :)
could you cheat and steal a front IRS off something? probably not as the geometry would be odd...
I have been thinking along those lines for the IRS. Probably using kit car parts as they are readily available and fairly cheap compared to one off's or the time and effort required to make an accurate pair myself. And they do tend to use Sierra hubs as well, so it makes brakes cheap too. It may not be the lightest set up if it were done that way, but it wouldn't be expensive either.
Oh, and I do have a very vague plan for the future of a really silly midget with maybe the high revving Mazda V6, so anything I do could always be improved over time for if/when I build something really special.
This thread was discussed between 25/11/2010 and 30/11/2010
This thread is from the archive. The Live MG Midget and Sprite Technical BBS is active now.