Projects '31 A FHC build (design?) thread

"A government crackdown ensued, and many suspected Luddites were convicted, imprisoned, or hanged." So I have to ask--Did Ned Ludd send you here? I know that we are stuck in the past, but quality and craftsmanship abound here. Thats why I still want to see those A inspired body designs. Give it up.

 

Thanks

The cylinder angle is purely a matter of clearance: they hit the rails if they're vertical. It doesn't matter, though. It's a simple matter to position the lower eye so the effective arm is perpendicular to the cylinder axis, which is exactly what I've done. Much more important is to get the arm length comfortably within the required range, especially as I have yet to concoct a Vernier adjustment detail whereby the respective leverages might be tuned.

As regards the four cylinders, that is, eight chambers, I think I'll need to do a sketch...

 

Some sketches to follow, in due course ...

 

Lol, sounds like all of my current projects... I think I've got one down though...maybe.

 

I can't escape the feeling i have that this design will have a De-dion type setup in the rear with a Mumford link. If this is the case then the powerplant of choice will have to have a low centre of gravity. This design study is a mammoth task considering you are utilising a ladder frame. I look forward to some more drawings.

 

DeDion, yes, but no Mumford: at this stage. That is, if what I've got in mind doesn't turn out not to work. Rearend options are wide open, but gravitating towards Jaguar centre.

 

Depends on how you look at it. I admire your ingenuity, but it appears overly complex for the performance I'd demand of a front suspension. Call my expectations low, but a dropped I beam, buggy spring and radius rods work for me. I can't speak for everyone here, but for me, much of the attraction of "traditional" (whatever that means) hot rods is the simplicity of their design and components. Less is more.

Bob

 

Here's a diagram of the hydraulic roll-control:

If a) both front wheels, b) both rear wheels, c) any diagonal pair of wheels, or d) all four wheels move together in the same direction, there is no change in pressure in either of the circuits. Only if both wheels on one side move in the same direction - i.e. roll motion - is there any change in pressure.

The pressure is resisted by the springs in the middle, one on each circuit. The little white squares represent damping valves. The front-rear proportion of reaction at the wheels is absolutely-fixable by the dimensions of certain components in the system: it can't change no matter how much the frame flexes.

The key to this is that you can't really speak of front roll stiffness or rear roll stiffness except as a proportion of the entire system's roll stiffness. The roll stiffness you'll find at either end at any time will depend entirely on what the other end is doing.

 

It´s like a hydraulic multi-way anti roll bar, or?
Clever stuff.

 

I hope it's clever! We'll see. But thanks.

 

I love stuff like this. To me it is
typically British (UK?). Overly
complicated to attain something
that can be done simpler. I look
at the 16 pivot points and think
that slack in them will negate
anything positive. Love the idea,
but I believe it to be way too
complicated to work reliably.
I say go for it. Let me (us)
know how it works.
Thank you for sharing and
good luck.
Matt

 

I hope it turns out less complicated than it looks! Thanks.

In mitigation let me point out that I've got three systems in parallel, each doing a different thing. There's the Mumford linkage to locate the axle, the air spring/bell crank system to support the weight of the car, and the hydraulics/bell crank system to keep the car upright (although the same bell cranks are used for the latter two). Each of these links the axle to the frame in a different way. One might therefore expect the lost motion in the whole to be coincident rather than cumulative, that is, equivalent to only one of the three systems.

In simpler language: Mumford links are known to work, e.g. on the back of Lotus 7 replicas and other things. Bell-crank suspensions are known to work, e.g. on Dan Woods T-bucket front ends. The two should work equally well together as long as they work in parallel, that is, as long as each system is complete in itself, from the axle to the frame, separately.

In another sense this approach is rather German: one function, one part; and no number of parts is too many for the Perfect Automobile. By contrast the French were always fascinated by intellectual economy, and would hold the entire car - all twelve parts - together with a single long bolt if they could. But here, separating the systems like this allows me not only to adjust and tune each one separately but also to keep them sorted out in my head.

 

I like it....

 

verrrrrry interesting............and after a few attempts of understanding I think it may just work..........What engine/trans are you planning on using?

Working on a 35 Auburn now myself.........

Mark

 

AH HA!!!!!

iv been thinkin of a setup identical to this for a couple of years now! Basically, your using the spring compression of body roll in a turn to force hydraulic fluid from one cylinder to another, thus dampening the amount of overall roll throughout the entire movement of the body, and countering the effect of body roll!

Pure genius! If i had the money, id have already developed it, but alas, too many irons in the fire. I always wondered what if though....

 

Because the project is about all kinds of other weird stuff I try to resist a weird engine, too. I actually considered a blown Audi five-cylinder for a while but, setting limits to which weirdness will have to be confined, it'll be a plainstraightforward SBC, but a stout one. I'm open to suggestion for a gearbox. My requirements are: manual, at least five speeds, divorced location alongside driver's knee. I'm pursuing a few lines of thought there. The final drive is gravitating towards a Jaguar diff unit.

 

Please translate this "stuff" from Dutch to English (or Spanish) so we Americans can understand it?!?

If ya call it "hydraulics" and give it enough pump and pipe diameter to HOP when ya hit the switches it'll be traditional!

How come the frame rails are parallel?
That looks more like a '27 T frame, with A horns added to the front.
Model A frames are an "A" frame shape all the way back.

 

I thought I did ...

No pumps, won't need them; but ½" pipes should give me the fore-aft response I want. It's a passive system, like (manual) brake hydraulics, only bigger.

Because I want them that way.
I know what a Model A frame looks like!
The parallel rails work with the underslung rear and all the other stuff I've got planned for the back of the car.

 

I like it also. i think it's an experiment that has some merit. All of the suspension and chassis designs- regardless of what you consider "traditional"- came from a learning process. Actually, it's called a Design Process, and it's what makes the cars that people on this board love- worth working on, improving upon, and experimenting with.

I'm sure that plenty of people told H Ford that he was misguided for wanting to mass produce cars on an assembly line. Mostly just because they were personally unfamiliar with the process.

 

As you may have guessed by now Dawie, I am a big fan of the same minimal is is often best type thinking.
I have done a fair bit a suspension modeling, and would tend to agree that being able to control the roll stiffness seperately to the vertical stiffness has some merit.
Whether the extra complexity ends up giving you an advantage over semi-fixed relationship systems is unknown.
Certainly from driving Citroens where the hydraulics plus big anti-roll bars are used to provide low vertical stiffness with high roll stiffness, I can say that improved roadholding, cornering etc is possible, but it's done using a very simple set of parts.
But then any arguement I might have about the shear number of joints is countered by looking at the back end of my GT40 which has 24 spherical joints to make it do what it does, that works fine, but playing arround with the set up is nightmare, far too many addjustments possible.
One thought however, why not just a panard rod with multipule holes in both axle and chassis brackets, so the roll centre of axle can moved verticaly, that will still allow you to tweak what the axle does with out a lot of complex bits tacked on ?

I build trikes,
effectively same as zero roll stiffness on one axle,
but given good wieght distrubution and well chosen instant centres,
still makes it round the bends at speed.

Years back a friend had a Ginetta G32, fast little car that cornered great,
110mph through sweeping bends one hand on the wheel,
Suspension, stock late model ford, with all the stock horrible rubber bushes.
Big eye opener that one !

Hey I have just found this real long bolt........

 

I'd actually looked at moderate to high roll centres for quite some time. Funny you should mention Ginetta in the same post. The early Imp-based ones worked quite well with high roll centres. The theory there is, get the roll moment arm as short as possible, get rid of roll moment around the roll axis, and you don't need a lot of roll stiffness. But various factors got me back to a low roll axis. Now, because I'm putting a lot of effort into splitting lateral weight transfer in a specific proportion front-rear, I don't want a lot of weight transfer coming through the roll centres - proportional to the weight distribution - and therefore being unresponsive to the roll-control system. It'd sort of defeat the object.

Have you ever considered a tilting trike on Ariel 3/Honda Gyro lines? The concept fascinates me. I'd like to see something like that with a V8.

 

I think that's the point, what I am saying is that a big chunk of the standard 'how to make your suspention better' thinking is based arround having to work with front/rear weight problems, tyre limitations, and variable ride heights, etc,
However take some of those out and things get better anyway, the G32 scored on the same trick, they sorted the fore and aft, and like magic the McPhereson/Chapman gives better results than any mod we could have made to improve the stock ford.

If you have got fore and aft under control so no issues there,
then most of the things which had you trying to design ever cleverer systems to get round the current problem go out the window.
Ie, you may be trying to trying to find a solution to a problem that is not going to trouble you as a result of you having solving the bigger issue already.

You will have gone for a nice low C of G,
so roll momment arm still going to be short compaired to stock production saloon.

Yeah have sketched a few leaning trikes,
but with all three wheels leaning rather than just the one.
Yes looks great on the drawing board,
but difficult to justify in terms of the hours involved/verses saleablity when finished,
I tend to be looking ahead to the next project at all times,
so that means what ever I put time in to has to be saleable to pay for the next one.

 

Ned, this is amazing stuff! I need to go back and study my suspension design books before I start asking stupid questions. I never really understood a Mumford linkage, but I'm starting to catch on... slowly.

I love the classic proportions of the FHC drawing in your avatar. I look forward to seeing more of this.

John

 

I've developed the front end a bit further.



A fellow HAMBer who asked for a certain reason not to be identified put me on to this simplification of the Mumford linkage, which also allows the front spring to be integrated with that system, saving me a pair of bell cranks. They'll come back when I sort out the roll-control, but they'll be smaller and simpler.

The spring doesn't appear here because I'd frozen the layer when I took the jpeg; besides the three-bar still has to be finalized, and the way the roll-control ties into that - and that leads right into the steering.

It's a process of paring, honing, polishing; thinking again and revising. Every new consideration sparks a rethink. In my experience this does not breed compromise and complexity but just the opposite. Things start to fall where they ought to be, and that allows simplification.

 

You know I am going to approve of the reduction in the number of parts,

I like the way the roll centre moves slightly in the same direction as the roll,
I have always thought that there is no harm in that.
( and even wondered about trying to introduce more of the same ).

Still not quite sure where you are going with this Dawie,
As I have said to you before, if the aim was better cornering then I would have just gone straight in with IFS,
Or if the aim was the look of old school beam, I would have just used a wish bone bottom link and very long radius arms, still gives you relatively low roll centre and good location, pair of coilovers and it will still corner more than well enough.

I like simple solutions to complex problems,
there is an elegance to that.

I have an old rover metro rear frame which has an A frame welded to it so it can be towed, has the usual hydro gas, linked left to right, I use it for transporting motorcycles, always amazes me that when I watch in the mirror the bike sits fairly upright no matter what bumps the wheels have to cope with.
The previous bike hauler was BMC mini based on un-linked rubber doughnuts, bike sat if anything lower, but pitched side to side more when hitting bumps.
I know none of that has any connection to building hot rods,
but it has made me question the whole roll centre, roll couple, roll stiffness issue again in my mind.
I also think back to the other thing we talked about before with the Triumph,
my 2500TC cornered better than the 2500S which had increased roll couple and a sway bar, lowering the spring rates and then trying to add roll stiffness back in did no favours, car was better off being controlled on just the springs, adding the bar just increased the understeer from tollerable to excessive.

 

I am anxious to see if this is a space ship or an actual car. I wait
patiently for photos.
Good luck

 

As you can see it's a process of going back, rethinking, following all the implications, rethinking again. It's a lot easier if you start somewhere and try whatever silly idea pops into your head to see what ramifications there might be - and be prepared to ditch it even if you've followed it for some ways. The important thing is to be clear about your basic concept, and to keep that ever in the back of your head.

I've been doing some of that. Having simplified the Mumford/air spring thing considerably I moved on to the roll-control system, to see if I could integrate that with the fore-aft axle locating components. Hence I switched to a side view:


I was briefly stumped by the logic here. The requirements for my three-link are:
1. The two symmetric links should be as long as possible, to limit roll steer.
2. The two symmetric links should mimic the drag links (twin drag links like on Foose's P32) to limit bump-steer.
3. The longer link/links should be at the bottom, for self-correcting anti-dive geometry.
4. The links that integrate with the roll-control should be short, there must be two of them, and they need to have some lateral freedom.

This had me going around in circles until it became obvious that the drag links should act on arms attached to the bottom of the spindles. That allows me to align the drag links with the lower links. I'd turned around at articulated drag links with idler arms, which would have added a whole order of magnitude of extra slop. Acting on the bottom of the spindles the drag links can be simple and almost literally straight-forward.

Now, I've still got one link too many: two upper links incorporating bell-cranks for the roll-control, and two plain lower links. Especially with that difference in link length, it's going to twist the axle at some combination of bump and roll, unless some adjustment is made. I can't lose a lower link, because both locate the axle right for the steering; I can't lose an upper link because I need to control roll in both directions. But both of the roll-control cranks don't have to locate the axle. If one were to incorporate a sort of shackle to allow a bit of fore-aft motion the other would be able to control the axle's caster positively without conflict. Hence, asymmetric bat-wings it is, and asymmetric cranks, one able to pivot slightly about a vertical axis and one not. The important thing is that they both give identical mechanical advantage.

(I think it was Enzo Ferrari who retorted to a customer who complained that his car had a slight dissymmetry, that one never sees both sides of the car at the same time ...)

I've added my BMW 728 radiator, suitably narrowed. That position, leaning forward at 30°, is provisional. I have not fully investigated the implications: though it does lower the top of the radiator enough to allow some direct cold-air induction to the engine through the topmost bit of the grille. I'll see if there's room for an expansion tank in there, too.

Now I'm ready to move to the rear suspension, though I'm sure I'll revisit the front - and everything else - more than once.

 

I find it quite fascinating although I don't quite understand it yet. Keep up the posts as long as it is tolerated here. Innovation is what many of the past racers and hot rodders were all about. If Doane Spencer were alive and posting his work he would probably get a lot of grief as well. After all what is "traditional"?

 

I've got a part!

This has just arrived, all the way from Zimbabwe:


And the back:



Some repairs will be needed:






The side frames came with it:


but I won't be using them, except as examples from which to fabricate toeboard supports. They'll give a good idea of the desired A-pillar profile, though.

So, at last, the '31 is rooted in extensive reality, if only a little.

 

Could be a long ole build then Dawie ?

I still like the diagram of the hydraulic system,
but then you know I am a big fan of oil under pressure.