VCUs – The sweet spot


Rather than highjack Hippo’s current thread on wheel size – I would like to put in my two pence worth on a new thread.


My Algebra is pretty rusty – and clearly when Land Rover first introduced the Freelander their’s was not spot on first time so no shame there then! – so I like empirical data – that’s why I like Hippo’s data collection projects – good fun stuff for people like myself who take an interest in our vehicles and how and why they perform – or break!


When it comes to VCUs and the loading on the other components there are a lot of factors and I think that Land Rover were on the money when they obviously deduced that by setting up the slight difference in gearing front to back they could reduce the load on the drive train – but the difficult thing that they got wrong on the first pass was the amount to hit the sweet spot.


I learned a lot about the characteristics of the VCU by doing Hippo’s wheel up test and comparing my results with others.


I think Hippo’s current quest for data on loaded rolling radius front to back will show that the fairly typical difference is circa 2% which is more significant in itself than the 0.8% and later 0.2% difference in gearing.


This illustrates I think that the fact that the assumption“If the front and rear diff ratios were the same there would be no drive at all to the rear wheels unless one or both of the front ones had lost traction and was spinning” is an over simplification – more accurate off-road where slippage is significant than on-road where actual slippage should not happen – but there is another factor - and that is slip angle – at least that is what it is called when considering oversteer and understeer brought on by the sideways movement of the vehicle brought on by tire deflection.


To appreciate just how much this can contribute just have a quick look at this link:-

https://www.tut.fi/ms/muo/tyreschool/moduulit/moduuli_10/hypertext/3/3_1.html




Which brings me to Hippo’s next experiment – which is looking at the forces when a VCU turns several times a minute – and Grumpygel’s example that at 100KpH and circa 2% difference front to back the VCU would slip 57 turns a minute.


I think the results will show that in practice IT JUST DOES NOT HAPPEN – the loads cause the discrepancy to be dissipated in other ways – notably tire deformation.


I look forward to Hippo’s results on the VCU loading – I suspect the loads to achieve 57rpm slippage would not be sustainable for long if that was actually happening.


So – what I am now thinking would square the circle for us empirical types – is another experiment – a couple of inexpensive electronic rev counters looking simultaneously at the prop shaft in front of and behind the VCU to see what is really happening at 100KpH on a straight flat road.



I have one and could afford another and if I can figure out a way to read them with a webcam or something to get a remote read I am tempted to build a rig and try it. (Any suggestions on this very welcome)


Just picture this scenario – and to put it into Hipponese “this un al urt ur ed”



Each little pair of tread blocks that come round looks at the forces acting on them and either push – or flex and do a little Michael Jackson moonwalk to comply – thus sharing out the load – and hitting THE SWEET SPOT.

Some may describe these VCU discussions as the thick leading the even thicker. If all they are going to do is state that they are so clever and people on the thread are so thick, they should just can feck oft and just communicate on threads where the others are up to their own (self percieved?) levels of engineering brilliance. If they want to contribute and constructively partake, then good on them.

What you've put up here is the sort of stuff that I like reading But I'll have to read that link again cos as you say - it "urt me ed".

Each wheel has an ABS sensor telling what speed they're turning at and they are connected via metal to the props - so there will be no slippage. This data can be gathered from the diagnostic socket and the diff ratio 3.2 applied to calc the prop speeds.

Have you calculated how much each tread block would need to flex to take up each slip rotation of the VCU?

I am not talking about actual slippage that an ABS sensor should pick up - merely the creep that will occur in response to a torque applied to a wheel in contact with the road - but Hey - that's a thought - if as you calculated there could be 57 turns/minute on the VCU at 100k with circa 2% difference - would the ABS be able to pick that up? My diagnostic dongle is pretty basic and would not tell me that - (ELM327 OBD11)
Never mind trying to follow the maths on that link - just think about how much sideways movement you can get by just pushing a car sideways - and remember the easy to understand descriptions of how that can generate understeer on front wheels and oversteer on rear wheels -they are easily grasped concepts - then think about the same effect only forwards and backwards - I think this allows the front wheel drive to get a little ahead of the back and only the difference need be taken up by the VCU - and the whole system shares out the torque generated and the resultant loads - Land Rover took an intelligent stab at this then modified a bit when they found they needed slightly different gearing to achieve the sweet spot. - And no - I haven't calculated how much - bit rusty - just embraced the concept - I'm sure you did too.

I think I'm gonna have to build that rig somehow to check the actual prop speeds front and rear.
In the meantime - Power to yer elbow Hippo (;-))

A diagnostic device like a hawkeye can read the wheel speed from the abs computer menu. Figure given is kph or you can change to mph on the hawkeye I think, if changing the view settings to metric. kph has a greater granularity but at 100kph 1% difference is 1kph. Preciseness of speed isn't as good as required as the figure given is a whole value with no decimal. Hence nearest value or "count to date" value per interval given as a kph reference.

You can film yer props with a camera if you mark the props with something to see the difference between them. High frame rates are preferred as 1 wheel turn = 3.214 rear prop turns.

My little revcounter just needs a piece of reflective tape on the shaft - two of them side by side and viewed together would do it - I need to rake through my junk to find a webcam or something to do this - actually while writing this I went on ebay to check how much my digital tachometer had cost (£18) and the first hit was on a hall effect tacho with a flying lead for £6.99 - I think I will risk a couple!!!!!!!!!!!!
Edit - 2 on order from HongKong £11.98 (don't you just love the net)

Having recently got hold of a 2003 TD4 that had a drivetrain with no nasty noises / behaviour (fingers crossed it remains so), I took the propshaft off mine temporarily as the vehicle had different brand tyres between the front and rear axels. Electronics is my job, so was thinking about making a small monitoring device that I could put near the VCU to monitor the behaviour of the front and rear shafts. A small embedded microcontroller, taking the input from either a photodiode looking at a reflective strip on the shaft, or a hall-effect switch sensing when a small magnet passes it. From this I can then see precisely what speed the propshafts rotate at, and establish when the load gets transferred to the rear. There are times when this is intended of course, but, if it starts to show up when not, i.e. a flat straight road, then I know there is a problem. The data can be used to simply illuminate an LED when the drive is being sent to the rear, or more usefully, stream the RPM of each shaft into the cockpit via Bluetooth and show it on a phone, where the distribution can be calculated too.

I will of course post results of the progress I make on this, which I will most likely do as the VCU goes back on after I've replaced the tyres. Which I want to do soon with Winter coming!

Jim

Nice one Jim.
Wouldn't it be great if a kit was available which monitored the prop speed front and rear, as well as temp of VCU, IRD and rear diff. Then we could monitor VCU condition at all times and know if there was a problem.

Now there is a challenge and an opportunity for you.

That's wot I just ordered the bits to do -I hope - a couple of low cost hall effect counters with flying leads to LED displays - looks like 4Bee4Bee might beat me to it - the more the merrier!

Even betterer, we've got a race to see whose first to get it to the market.

There's already someone at the finish line reading the abs sensors and both props.

yu dont need to have two rev counters - you only need to monitor the difference between either side of the VCU, not know absolute values.

The Mad Hat Man said:

yu dont need to have two rev counters - you only need to monitor the difference between either side of the VCU, not know absolute values.

Click to expand...

I think they're looking at doing x - y = speed difference
Yer need to know both prop speeds as one it a datum to measure the difference of the other.

Yeah - differences should be small so simultaneous reading is called for - also would let us look at results under different load conditions.
Hey Hippo - if anyone has got data from a hawkeye could they publish here to give us a benchmark - I know you said it would only be to I K/Hr but it would be a start

Prop speed when turning is interesting anorl.

Been thinking about this.
Is it likely the prop speed difference would change with a stiff VCU? I thought the whole problem was it didn't change when the VCU stiffened up causing something else to give, like IRD or diff.
The only way the prop speed difference could change is if the tyres were to scrub and on dry ground is that likely?
I guess that is what these two threads are all about.

Alibro said:

Been thinking about this.
Is it likely the prop speed difference would change with a stiff VCU? I thought the whole problem was it didn't change when the VCU stiffened up causing something else to give, like IRD or diff.
The only way the prop speed difference could change is if the tyres were to scrub and on dry ground is that likely?
I guess that is what these two threads are all about.

Click to expand...

very true,unless the v/c can cater for the difference stress will be placed on the diff ird and tyres etc

jamesmartin said:

very true,unless the v/c can cater for the difference stress will be placed on the diff ird and tyres etc

Click to expand...

That's it in a nutshell - the whole setup is a compromise - things like badly matched tyres are known to compromise this - a few decent speed difference measurements and some torque figures at given revs from Hippo's next rig would be an interesting insight

(for us mere mortals with merely an intelligent interest and not perhaps having the ultimate faultless knowledge of - for instance - the Sunday poll winners)

In theory the vcu should be able to take out any differences (absorb if you like) in normal driving conditions.
That would include straight line driving to tight circle driving.
Tight circle driving is where the prop speeds differ the greatest.
Hence peeps first suspect something is wrong when turning starts to feel different.
It's my view that as the vcu starts to "stiffen" up over time (or whatever factor) straight line driving which only puts a small stress on the vcu with minor prop speed difference isn't effected to such a condition where it's noticeable.
But tuning I tight circles or reversing round a corner does feel different as the vcu has stiffened and fights more against the props needing to turn at the greater end of the expected speed difference.
The mechanicals can't lose the additional force not absorbed by a good vcu so it's either down to tyres slipping or something mechanical braking. Hence ird and/or rear diff cracking.

Hippo said:

In theory the vcu should be able to take out any differences (absorb if you like) in normal driving conditions.
That would include straight line driving to tight circle driving.
Tight circle driving is where the prop speeds differ the greatest.
Hence peeps first suspect something is wrong when turning starts to feel different.
It's my view that as the vcu starts to "stiffen" up over time (or whatever factor) straight line driving which only puts a small stress on the vcu with minor prop speed difference isn't effected to such a condition where it's noticeable.
But tuning I tight circles or reversing round a corner does feel different as the vcu has stiffened and fights more against the props needing to turn at the greater end of the expected speed difference.
The mechanicals can't lose the additional force not absorbed by a good vcu so it's either down to tyres slipping or something mechanical braking. Hence ird and/or rear diff cracking.

Click to expand...

So round a bouts are a challenge for them?

dfossil said:

That's it in a nutshell - the whole setup is a compromise - things like badly matched tyres are known to compromise this - a few decent speed difference measurements and some torque figures at given revs from Hippo's next rig would be an interesting insight

(for us mere mortals with merely an intelligent interest and not perhaps having the ultimate faultless knowledge of - for instance - the Sunday poll winners)

Click to expand...

I'm hoping to be able to say my new gkn vcu (it's had some use after tested when new, then tested after some miles) can turn say N times per minute. If the same test is done to a worn/suspect vcu and for eggsample the result is N/3 (one third turns of N) then we may have a rough guide to work too. I can't do magic but I think this is possibly the way forward for use to be able to approve or condemn a vcu as fit for use.