I've been very busy at work this week, so I haven't had a chance to do much.
However, I have done a little bit on enhancing the data collection sensors that the car has. First up was an steering
angle detector. You can buy expensive (about £150) string potentiometers to do this with but I'm rather more
of a cheapskate. So, I constructed this device which is made out of a 10-turn pot and a pulley and belt nicked
out of Tom's ancient Lego Technic box and modified slightly to fit the pot's shaft.
First indications are that it works OK. The obvious problem is the belt slipping. If it does then I'll just use a different material and, possibly, double it up around the steering column.
Next, I've been wanting to make a detector for the lap timer
beacon for some time as I could feed this into the DL1. This would give me two advantages; first of all I could
get a proper marker on the DL1 track maps as to where the start/finish line was. Secondly I could use the Palm
to display the lap times and therefore lose the lap timer box.
So, I made this little device. (It is little, those holes are on 0.1" centres.) It's just an integrated IR detector connected up to drive the beacon detector input of the DL1. The only issue is whether this is the right sort of detector to respond to the beacon at circuits. It certainly detects a domestic remote control, as does the existing lap timer, but it's possible that the circuit ones are different. The only thing to do is to wait and see. Just in case it does work I installed this little thing in a box and potted it with some epoxy. That should make it work even though it's out in the elements. Rather permanent though...
Next thing is to write some software for the Palm to actually display some times. I'll have to enhance the software's user interface quite a lot to do this though, something that I've been trying to avoid, just because of the time it takes.
I'm really going very slowly at the moment, but I guess it is summer, although that isn't
really clear from some of the weather we're having. I'm now working towards going to Oulton Park next week for
my training day with Tim Harmer. As such I'm really just putting the car back together.
In an effort to fix the drift that I see in the output from the yaw-rate sensor I built a rather more sophisticated interface for it based on the circuit that's in the application note for the device. The version before just stuffed the output and the reference output directly into two inputs on the DL1. This one includes an op-amp to condition things a bit. (The op-amp is the right hand one of the chips on the little daughter board in the photo above.)
At first sight it does seem to have improved things slightly, although I won't be able to tell until I actually run the thing for a while and look at the logs.
As part of getting the car back together I thought it was about time I took off one of the covers and had a look at the fuel filter. If you remember, this is a filter that I made from hybridising two different sizes of filter and it was indeed slightly grotty. I've got some spare elements so that's easy to fix. However, in doing this I noticed that there were some very small cracks in the plastic cover. They don't actually seem to go all the way through but it would seem sensible to replace it anyway. So, I've just ordered a new one from Demon Tweeks.
I've been wanting a tyre rack for the trailer for a while as carrying a spare set of wheels and tyres
is both necessary and a bit of a pain. So, having phoned Brian James for the price of a kosher job and then picked
myself up off the floor I finally decided to make one.
So, I bought a load of steel from MacKay's and made this thing you see here. The tyres/wheels just sit on the ledge and you have to tie them down with luggage straps which is a bit of a fiddle but as it cost me less than a fifth of what the BJ version would have cost it seems pretty fair.
I'm quite chuffed with that, as you can probably tell.
So, with that done it was off to Oulton Park for a day to be spent hammering
around the track with Tim Harmer.
We spent the evening before the trackday on a camp site local to Oulton. God knows what the various caravanners, stuck in their little metal boxes watching their TVs thought of Tom and me arriving with a race car in tow. Whatever, we got to the circuit early the following morning and met up with Tim and signed on.
There was another RGBer there in that Doug Carter, who drives a very neat class A Genesis, the twin of the car that Tim used to win the championship last year. In fact, Doug was in the garage next to us.
However, Doug sufferred a failure of an oil union and a small fire as a consequence which curtailed his day. he also dumped oil all over the track which was a pain, especially as we nearly ended up in the gravel trap at the hairpin. Fairly hair-raising that. Doug seemed quite hacked off about the problem but it's really just one of those things.
After signing on and driving a few familiarisation laps we got Tim fitted into the car and I went out for a first blast of the day. Frankly I drove horribly and I was embarassed to hear what Tim was going to say. Luckily he was very kind and suggested a few things and we went out again.
This time things were a lot better and I started managing to get the power down rather more forcefully. The biggest problem though, was Lodge which was pretty much the bugbear when I raced here last year. It's the corner at the top right of that map (which was produced by the DL1's GPS receiver).
After another couple of runs we stopped for lunch. I did seem to be getting a bit faster, and looking at the datalogs later this does seem to be the case. One odd problem was that the gearchange seemed to have moved in that the actuator arm behind the steering wheel was no long parallel with the wheel, which was odd and I put down to something having slipped somewhere. I adjusted it slightly but it seemed a bit odd all the same.
After lunch we decided that Tim was going to have a go at driving the car so we swapped seats and rammed a load of padding in behind Tim. He's not a small bloke but he was amazed that when sitting in my seat he could only reach the pedals with the tips of his toes. Sitting in the passenger seat with Tim driving was an amazing experience. Essentially the bloke's an animal and he was much more brutal with the car than me. In fact, my fastest lap was actually a good deal quicker than Tim's by about 2 seconds but it was clear that he was indeed doing the sorts of things that I've realised all too well I need to do. However, one issue was that he didn't seem to be amazing earlier onto the power than me, which is one of the things that I was most concerned about. (At least, this applies to my quicker laps, the slower ones which are probably more representative of the times I put in at Brands a while ago do suffer from this.)
As an example of the differences between me and
Tim, here's one simple bit of the logs. (Click on this to see the graph in a separate window at a rather better
resolution.)
This is a trace of throttle position around the Old Hall section of the track. The black trace is the values from Tim's best time, the red trace is that from my best time (which was 0.09 seconds slower than Tim's time). Finally, the green line is the DL1 calculated "time slip" representing the amount of time that I've lost compared to Tim.
Obvious things to note are:
- I've got out of the throttle ages before Tim. In fact about 30m before him which is an absolute age. (So much so that I think someone must have been in front of me here. Problem is, this is my best time in this sector.
- Tim gets out of the throttle and hits the brakes hard, whereas I'm much more gentle about it. Mind you, as he's 30m down the road he's probably scared by now.
- We both actually get into the throttle at about the same point. However, Tim is much more aggressive but he has to keep backing off to control oversteer (which can be seen on the yaw rate graphs which I haven't included here.)
- I actually get back to full throttle a bit before Tim, which is pretty astonishing. This may be a consequence of a much gentler application of the throttle allowing the car to stay a bit more balanced.
- Tim actually goes around the corner faster than me, which is why the time slip is up to over 0.2 secs in the middle of the corner.
- As a consequence of getting on the gas earlier I claw back time from Tim on the run down towards Cascades.
- However, my peril sensitive right foot doesn't like the blind brow and I have to speak sternly to it to make it stay down. Without that I might even have been faster through this section which, considering where I started slowing down, is pretty amazing.
Having spent a while looking at the logs now I'm getting quite pleased with the day in that there's lots to find out here. It does show that I should be able to get there.
One problem, though, is that my best time as 5 seconds slower than the best time in the race last year. I'm hoping that this is just down to the extra weight of having Tim in the car. What was incredible was just how quick he was "out of the box". I'm confident that with a fe more laps he would have been way quicker than me, which is sort of good really.
Biggest problem though became apparent while Tim was driving. He was having
more and more difficulty changing up, especially on the way out of the hairpin. Eventually we came in and when
investigating realised that as you pulled the gearchange lever the engine moved which was rather alarming.
On investigation we discovered this which is a failure of the main engine mounting on the right hand side of the engine. This is what the mounting originally looked like, where the bit that goes to the bottom right of this poor photo is the part that points towards you in the clean photograph.
Clearly, the mount has failed due to stress. This is probably a consequence of vibration which has been a problem in the past. What I shall have to do before the Lydden race is weld this mount back together. I'll probably try and weld it simply, and then add some extra gussets for a bit more strength. (More weight...:-()
I managed to get the engine mount off the car by undoing the other mounts and
jacking/shoving the engine across. With the bits in my hands it was clear that the mount had failed in the expected
place: at the edge of the HAZ around the weld. The missing lump of steel in this photo is still firmly attached
to the leg itself which is the bit at the bottom right of the previous photo.
It would have been nice to have the time to complete re-engineer this mounting; I've thought for a while that something rather more like the mounting cradle that appears, for example, on bike-engined westfields would be nice. However, for now I'm just going to have to bodge something together.
So, I rewelded the leg back on and added a couple of strengthening (I hope)
gussets. The mount then looked like this. Looking at the photo now I should have added something between those
two gussets as well.
I guess I'll just have to keep a careful eye on it in the car itself. Once back
it it looked like this and fitted back into the original mounting holes with the minimum of bodgery. Looking at
this, I would really like to do something about these, I'll havea think about it once the close season comes around.
With that done the car's reallly ready to go to Lydden this weekend. However, I shall spend some time checking all the bolts and inspecting things like brake pads. I also changed the oil yet again, too.
It'd be nice to clean some of the mud off the car too, the underside looks like a field at the moment.
