Fake - F7R Motorsport Daily

[Chris_C]

Cheers gents. I intend on doing these bearings rather than swapping the axle when the time comes, but Fake's had done 212,000 as the big pile of receipts didn't show anything for them.

The main reason for changing the axle was I wanted slightly more rear brakes, and Jason had this in the shed. It made financial sense as everything one it was new

Steve, yes, it makes that amount of induction noise (it's quite at motorway normally, but that is a relative thing, there is no noise until second throttle, then all hell breaks loose). No exhaust noise as I'm still running a stock 1.7 exhaust.

[Chris_C]

Not updated this for a while.

Recent things... Steve_P managed to find me another set of the silly rare Ronals so I could replace the one bent by the flipping great pot hole.




So, I now have 7 and a half of those

I've replaced the rear axle on Fake with a 360 one, the wheel bearings needed doing anyway, and I've wanted the larger brakes for a while. No pictures of the swap as it was a bit rushed (started 6 hours before I had to be 200 miles away...) but the rebuilt axle is a heck of a lot nicer!

First couple of rallies have been done with the F7R in, and first few teething problems found and solved. Adam had replaced the silly springs on the clio downpipe (used to clamp it up to the manifold) with some ali spacers, which gave a much better clamp. However, it seems I may have been using a few too many beans, and in the second rally she suddenly got very loud, a check at the next marshal point showed that I'd lost one of these spacers. Continued to drive round (slowly, and away from as many houses as I could manage!) and *just* managed to get a win on the event. I bodged it up that night with a socket from the toolbox replacing the spacer, and did a proper fix the next day. It seems the ali had too low a melting point, and with the extended hard driving had melted the spacer loosening the nut and loosing it completely. The other side had been folded over like a tophat.

New spacers are now made from steel, and Aerotight (high temp nyloc for want of a better description) nuts are on order.

Finally, I've been fiddling with Renault ECUs. The ChrisChip is in the break period, whilst I try and forget about the design enough I can go back to it in a few weeks to see how many silly mistakes I've made. In the mean time, Tommysb has sent me a blank 27c256 EPROM, and I've replaced the random chip that was in the car with a geniune Williams map.

So, first things first, was to take all the ECU's and EPROMs that I have lying about on long term loan from people and a few scrappy ones and stick them through the programmers to see what they actually look like. Some really interesting results, I'll include a couple of graphs in a sec.

The 'new' map is a much more refined jobbie than the one it's been on (thats not hard really, a smack round the face with a brick is too). It doesn't have quite the power of the old chip, but still isn't a slouch, but a lot of side effects have gone. Pinking obviously, as thats directly related to ignition mapping, but the idle speed is now correct (eprom doesn't control the ICV as far as anything I've read says), and the speed sensor working is more pronounced (with slow speed no throttle making the car put on 300 revs or so). There is no knock whatsoever from reset ECU, unlike previously taking 150miles to get to a happy retarded state with the old one.

OEM Fueling.

Randomchip Fueling


This doesn't make a huge amount of sense to me atm, the ebay one has smaller injector openings for all parts of the map apart from the high revs no load corner of the map.

OEM Spark

Randomchip Sparks


There is no advance (0 degrees) for the randomchip map at the lowest vacuum (not easy to see in the above graph, the leftmost line). There is less advance at the lowest rev range, and only a huge difference in the mid load area, where there is a change from 28degrees to 40.

Lots to be thought about for how I want to change it, and lots to learn to understand a few things in the above.

I'm interested to see fuel economy, I was returning 115-135miles to £20 for the last 2500miles, no matter how I drove it, or on what terrian. I'm now very certain there is much to be had from ECU fiddling, but not until I have a wideband and a good knock detection circuit. The thing I don't understand is the Williams map is lacking in variation, and also fuels more for low down power so in theory should be slightly worse if anything for economy?!? Unless I havn't got my head round it all properly, which is likely. Certainly I'm not convinced I'll get the petrol saving I was hoping for.

[tommysb]

How are you planning on doing knock detection chris? I have had a few ideas myself. Maybe look at some megasquirt stuff, as people have made knock detectors for that.

[Chris_C]

Me and Andy made a design for his squirted car a fair while back that never got finished. It's just uC coding that needs to be finished off on that, the boards are sat assembled somewhere in my projects pile. It's a Texas knock DSP that does the interesting stuff on it.

[foggyjames]

Thing you will find: fuelling requirement increases far more quickly with load than with RPM. Still makes very little sense to me, but that seems to be the way it is. The Williams map looks more sensible, to me. It seems to jump up a very long way off-idle, though. Get a wideband, and get experimenting! A set of det cans (i.e. a stethoscope and a length of pipe) will go a long way when it comes to listening for knock...

The spark tables look quite similar, except the chipped one has a load more advance at high RPM / load...which makes sense for making power with high octane fuel. Might explain your constant pinking once your toe is down. Whether or not it's safe is, of course, another matter!

cheers

James

[Chris_C]

Thing you will find: fuelling requirement increases far more quickly with load than with RPM. Still makes very little sense to me, but that seems to be the way it is.

Hmmm, interesting. Not what I expected at all.

The Williams map looks more sensible, to me. It seems to jump up a very long way off-idle, though. Get a wideband, and get experimenting! A set of det cans (i.e. a stethoscope and a length of pipe) will go a long way when it comes to listening for knock...

Wideband's not going to happen for a fair time, I can't afford it Hence why I'm down to base map until I can get my head round everything ans do a proper job.

[foggyjames]

I've not got around to watching the actual injector timings, but certainly in MS the unit used to determine fuelling "volumetric efficiency" doesn't double. However, it may multiply that by RPM further along in its algorithm. That looks very similar to an MS map, so perhaps it works the same way? Or perhaps engines are more weird than we thought, and they don't really use double (or close to double...I know they're non-ideal) the fuel at double the RPM. I suspect it's not the latter, to be fair...I think it's the way the EMS expresses it.

Got a laptop? From memory, an Innovate LC-1 is only about £100. If you a gauge, it's a bit more spendy, but meh...

cheers

James

[Chris_C]

I hopefully am borrowing an LC-1 from a mate next week. I have a laptop, and I've nearly got full OBDII working in the thing so I'll be able to get full data out of that, inc narrowband values, and knock amount. With that, and wideband, the plan is to log *everything* drive it around for an hour, get home, and write some thing to mod my map. Drive round again, repeat until map is good. Seems a better idea than trying to remap on the fly, my home PC has a heck of a lot more power in it and I can think about what the map looks like before it gets run.

[foggyjames]

Very sensible. You can reasonably map fuel on the fly with an assistant, but spark is trickier. Logging and retrospective mapping (for the want of a better way of putting it) is pretty much the only way to do it when flying solo.

cheers

James

[Chris_C]

First tank of fuel through (99oct). She did 99miles on just over 20quid, which is worse than before, but it was all mild-medium lane work so would have been in the biggest area of change in the fuel map (i.e, that big lump of red). She felt sluggish on the new map comparitively though.

So, I've chucked in a tank of semi skimmed (to see if she'd not pink on that more than anything), and she seems a heck of a lot better, pretty much all the pull of the old map is there. Now, I can put the pinking I had originally down to leanness, the complete lack of fuel under load in the old map is frankly scary. I originally suspected too much advance but after thinking over these tables I'm not so sure. I'm going to try and baby this tank around, and see what I get MPGwise if I can keep it out of that red zone.

That all makes sense in my head, apart from the fact it felt sluggish on 99 octane. I can understand it feeling the same, if a car isn't mapped for it then it won't make a difference, but why would it feel worse? Unless theres some random ECU adaptive bollocks going on, but the effect was really noticeable before and after the petrol station, and the tank was very low. Hmmmm. Need dyno results really, which are difficult when the old map needed 150miles to stop knocking and the fuel change is required, makes getting back to back runs near impossible. The car still has all the wiring for powerdyn though.

Narrow band display sourced (I think I'll choose the deluxe one over the standard, £8.95 rather than £5 though) and even though it's not super useful, it's a sexy set of LED's to watch. Further diagnostic interfaces were made today so hopefully she'll be fully OBDII complient soon.

[foggyjames]

I can't think of a good reason why you'd *lose* power with higher octane fuel, unless it was bad in some way. The only thing I can think of is that with some combustion chambers (Honda B-series, for example) it's possible to advance the ignition timing 'too' far (so the engine starts to make less power) before you hit the ping threshold. If the ECU was doing that automatically...? Seems unlikely, to be honest.

Without a wideband, you're pissing in the wind. It might be losing power by being too rich (big red section...although it has to be *very* rich to actually lose power), but you just don't know. It could be timing related. It could be that the 'butt dyno' needs calibration I say sort your fuel table out at the top (12:1s on-load) and worry about timing once that is set.

Narrow band display sourced...even though it's not super useful, it's a sexy set of LED's to watch.

That's £9 you could have put towards something useful...i.e. a wideband

cheers

James

[Chris_C]

Narrow band display sourced...even though it's not super useful, it's a sexy set of LED's to watch.

That's £9 you could have put towards something useful...i.e. a wideband

Agreed Thing is, and this is going to sound either very silly or very strange... you wouldn't believe how much fun I'm having going out in the morning, turning the key, and the car starting. And driving it without fear of something breaking. I've replaced that entire bloody car pretty much in the last 18months, and she now works.

I'm not spending ever hour I have spare trying to make sure I can get to work tomorrow morning, and I can actually do stupid pointless things again, and it's ACE. I'm really tempted to get the blue neons out again

[foggyjames]

I know what you're saying. I think I bailed on the 240 just when I'd fixed (with plenty of help from Andy!) pretty much everything that was wrong with it. Having to work on it almost every night once I got in from work....just to get it ready for work the next day...became tiresome, fast.

cheers

James

[SteveP]

Tis a nice feeling Although I think I've gone too far that way... and haven't really had to do any car work as such in the past year apart from the usual servicing which takes about 30mins! Best balance I found between tinkering and reliability was a 360 injection

Good work mate, must meet up soon - I want a spin in that thing

[Chris_C]

For future reference, if anyone wants to put a 360 axle on a 340, the most sensible way to do it seems to be late 340 front cable, then the adjusty thing, then any set of late 360 rears (2 lefts, 2 rights, 1 left and one right). You have to make a new bracket though. Will explain more when I have some more time, just adding this so I remember later

Fake is still *just working* (tm). She's been racking up miles at a rate of pace. It does seem that with great power, comes great fuel consumption. I'm returning 25-30MPG on the Williams map, about stock GLT return tbh. And for the *just working* ness, like a GLT, I'll happily pay it until I have time to look into remapping, she drives like a 1.7 on steriods, is as reliable as a GLT and is awesome. I've not been this happy with Fake for.. well, ever really.

Apart from the rallies I've done with it on this map. That was about 15MPG and less power than the old map. At least the engine doesn't sound like it's about to explode