21 Turbo G788 WUD

[Ben12345]

Right... well new caliper arrived and is fitted, bled and handbrake adjusted. Rear shock nut/bolt removed and a fine thread type one put in (m12x1.25). I had to cut the original one off way back when I had the rear subframe on the floor to do the welding work, and I replaced it with what I had to hand at the time, which was a good old m12x1.75.
I turned down the mixture pot on a wing and a prayer to see if it did anything, and the exhaust doesn't stink anywhere near as bad now. My CO meter is goosed, so it'll have to wait until the MOT for the final tweak to 1.5%.

And finally, MOT retest is booked for tomorrow afternoon. Hopefully he doesn't do a full test, i'd worry he might be looking harder and find something else wrong.

 

[DaveL485]

the final tweak to 1.5%.

If you want to bring it down a tiny bit more, block off the one way valve that joins to the inlet manifold, from the crankcase breather. That reduces it about 10%. Obviously turn the mixture pot down all the way which you already did and I think I remember way back when, if I disconnected the air temp sensor on the plenum chamber it reduced the CO down even further.

 

[Ben12345]

Hopefully it won't get that bad, but thanks for the tips. The level was at 6% on the original MOT, and the pot was almost exactly mid way before I gave it a twiddle.

The coolant sensor ohms out ok according to the book, but I didn't check the air inlet sensor. I'll have a look at it when I'm home at lunch time, just for peace of mind.

 

[Ben12345]

Finally passed!

Ball ache with the emissions though, glad Dave passed on his tips in the end!

Pot turned all the way down, still 5ish%
Disconnected air temp sensor, 3.65%
Clamped up breather hose, down to 2.5%

MOTer didn't like the breather clamped, so he suggested turn up the idle slightly. Raised to approx 950, and we got it down to 3.3%.

We pulled the vac hose from the FPR, and there wasn't really any change in the CO, so perhaps that's the reason. Would that make it so high though? Maybe that and leaky injectors i suppose. What's peoples opinions?

Cheers

 

[DaveL485]

glad Dave passed on his tips in the end!

Been a few years since I used those tricks, now I just turn up, flip the laptop open and watch my MOT guy roll his eyes as I turn down the master trim to get it below the threshold, then turn it back up as soon as he disconnects the exhaust probe hahahahaha

Great news though mate, happy for you!

 

[DaveL485]

We pulled the vac hose from the FPR

If anything that would increase the fuelling/fuel pressure as it's a rising rate reg, driven by manifold pressure. It would go from [2.5 bar - manifold vacuum pressure] to 2.5 bar.

 

[Ben12345]

If anything that would increase the fuelling/fuel pressure as it's a rising rate reg, driven by manifold pressure. It would go from [2.5 bar - manifold vacuum pressure] to 2.5 bar.

Yes, we did it to see if the fuel pressure regulator was doing anything... we were expecting to see the exhaust go richer, which it didn't. So maybe it's not doing what it's meant to.

I'll stick a pressure gauge on there at the weekend to see what's going on with it.

Another problem after driving it on the road... boost doesn't seem to be controlled. The in cluster gauge goes all the way to the end and it appears that the overboost switch is doing what it's meant to. Checked the amal valve, and it seems like it's plumbed correctly. That's about as far as I got though with my checking because of the dark, wind and rain. Another job for the weekend!

 

[DaveL485]

[Stolen from one of my old 21toc posts]

Standard boost is 13.2psi, or 900mBar, but in the std setup the actuator is set to open at 8psi.

13.2 is attained by a pulsating solenoid valve, AKA the amal valve. It pulses at a constant frequency of 12 Hertz, so <time open>+<Time closed> equals <constant> but it varies time open and closed to make the adjustment. It basically acts as an electronic bleed valve....boost controller stylee. When you are in 1st gear (2wd only) or engine temp is below 50 centigrade it limits boost to actuator pressure, ie 8psi, otherwise it changes its pulse profile to bleed away pressure and allows 13.2psi.
On top of all that, it also controls pre throttle butterfly boost pressure according to how much throttle you have on. Its a brilliant innovation, half throttle=half boost, instead of a setup like my old ph1 with no amal valve which hit over a bar of boost at 1/3 throttle. Great for fuel saving and cruising, prolongs the life of the turbo as it doesnt work as hard as well.

All this though, is assuming its working correctly. Problems begin when its not, from boosting low (8psi) to massive overboost (BOOM!). The overboost problem is dangerous but Renault saw fit to add an overboost protection system. It cuts in around 18psi and it breaks a circuit and turns your engine off. Again assuming its working, and hasnt been f**ked with by some boost hungry little scally. DO NOT REMOVE IT. Take the time to adjust and reset it to the required level.

If you do get problems with boost the first thing I suggest is eliminate the electronics, which in cliffs case has proved that the mechanicals are working correctly, so the fault lies with the electronics. Replacing the valve should prove fruitful, if not there is a possible wiring problem as earthing one of the amal valve wires results in it spazzing out and jamming open giving up to about 21psi.

 

[DaveL485]

....so run a pipe from the boost feed direct to the actuator.

If you see 8psi boost and no more, the mechanical parts are OK. If you see more, then its a mechanical issue with actuator, wastegate, or the pipework.

 

[Ben12345]

....so run a pipe from the boost feed direct to the actuator.

If you see 8psi boost and no more, the mechanical parts are OK. If you see more, then its a mechanical issue with actuator, wastegate, or the pipework.

Thanks Dave, It does seem as if the amal valve is knackered or the electronics. I pulled the hose off the valve that leads to the actuator, and put this on the boost feed and the in-cluster gauge now goes to about 11 o clock at full boost.
Now i've got an excuse to put in my greddy electronic boost controller


Back to the emissions though.... I put my fuel pressure gauge into the fuel feed line, and pressures were perfect. Just under 2 bar at idle and 2.5 with the vac hose pulled off. I guess that's not the source of my rich idle. I let the car sit for a few minutes with the engine off, and the pressure didn't drop, so the injectors aren't leaking either.

Next step... see if the MAP sensor and throttle position sensor are working right I suppose.

 

[DaveL485]

goes to about 11 o clock at full boost.

Yep, thats correct, 8psi. Standard boost is the 2nd mark before the 'red' area. You can go up to 16 or 17 safely. The valve works with a switched earth at the ECU to pulse it at the 12 hertz, but if its earthed out to the chassis it gives the massive boost increase you describe. Easy enough for a man of your talents to check for that permanent earth and fix the control feed back to the ECU if its that.

For rich idle the first place i'd look is the coolant temp sensor in the stat housing. The turbo sensors and the non turbo sensors look identical but operate in totally different ranges, so if someone has fitted the wrong one in it's history the ECU will get duff data. Or, the original might be goosed.
You'll have to check but I think the turbo sensor operates at 1000 to 3000 ohms and the non turbo one at 100 to 300 or something like that. Pretty sure the values are in the haynes but this is my mapping from the Adaptronic for the OE Turbo sensor (albeit a 27 year old one!)

 

[Ben12345]

Yep, thats correct, 8psi. Standard boost is the 2nd mark before the 'red' area. You can go up to 16 or 17 safely. The valve works with a switched earth at the ECU to pulse it at the 12 hertz, but if its earthed out to the chassis it gives the massive boost increase you describe. Easy enough for a man of your talents to check for that permanent earth and fix the control feed back to the ECU if its that.

For rich idle the first place i'd look is the coolant temp sensor in the stat housing. The turbo sensors and the non turbo sensors look identical but operate in totally different ranges, so if someone has fitted the wrong one in it's history the ECU will get duff data. Or, the original might be goosed.
You'll have to check but I think the turbo sensor operates at 1000 to 3000 ohms and the non turbo one at 100 to 300 or something like that. Pretty sure the values are in the haynes but this is my mapping from the Adaptronic for the OE Turbo sensor (albeit a 27 year old one!)

View attachment 89871

Thanks Dave,

I checked out the coolant sensor, and compared it to the values in the workshop manual, and it's about right. The air intake sensor is also looking good, as does the MAP sensor. Got about 0.8v at idle, and low to mid 2's with the ignition on but engine off, so it looks ok.
I measured the throttle position sensor, and it looks like it's working too, but it does look like someones been at it before and messed with the position. The book just says something about the expected value when using the XR25 diagnostic unit, rather than set it to x ohms at idle and x at full throttle.

Looking closer at the values specified in the manual, this may be the output values from an 8 bit ADC, with 255 being maximum and 0 being minimum, with the throttle potentiometer being wired at 0v on one side, 5v on the other, and the wiper inputting to the ADC. If so, I can do a bit of maths and get the resistance (or voltage) I need for at idle and the resistance for full throttle.

Assuming that's correct, then I would expect to see pretty well close to 5 kohms at idle, as it's a 5k pot, and close to 0 ohms at full throttle. I have nothing like this... well at least not at idle. It was something like 3.5 kohms at idle, so way out.

What do you reckon?

 

[DaveL485]

Here are two TPS mappings from two different cars on standard TPS's, @la21t and @dip21



Here is mine, which is converted to a conventional/universal aftermarket item...I didnt realise until just now they work the other way around to OE



Obviously the resistance is a linear increase between the 0 and 100 points. You are right in assuming that the 3 wires are power +5v, earth and signal, marked A B and C on the plug. Can I remember which is which? Of course not LOL

 

[DaveL485]

Check your knock sensor too, is it connected? And it shouldnt be too tight in the head, finger tight plus 1/8th turn.

 

[Ben12345]

I think the problem with the rich idle was the throttle pot.

I adjusted it by voltage rather than resistance in the end, because that's what the ECU is looking for.

From the workshop manual, it states that I should see a reading of 5 to 15 units at idle, and a minimum of 225 at WOT if I had an XR25, which doing a bit of maths means 0.098v to 0.293v at the output of the pot for idle, and 4.39v for WOT.

Before I adjusted it, I had about 0.35 volts, so outside what the ECU expects at idle. I went with about 0.13v, so closer to the bottom end of the limit. The idle is definitely smoother now, and it recovers much better after blipping the throttle too. The exhaust also stinks less. I'll have to either go to the garage to get him to put his probe up the exhaust or buy a new CO meter myself before I can find out for sure.

At WOT, I also have about 4.9 volts, so above the minimum limit of 4.39v that I calculated.

I had a bit of a play with the injectors after that, I have some Focus ST170 injectors, which are meant to flow about 1lb/hr more than the originals, and more importantly are the newer 4 hole type that's meant to give better atomisation. Well, they didn't work! The engine fired up for about 2 seconds before giving up. After a bit of googling, it seems that these old ECU's are smarter than I thought. The new injectors are high Z whereas the originals are low Z. The ECU actually measures the impedance of the injectors, and if it's not as expected, it'll shut down, which it what was happening.

There is such a thing as a 'renix injector emulator' which is about 30 quid, which tricks the ECU into carrying on when the injector load is not present. They are made primarily for the LPG conversion market, where the petrol injectors are turned off, and the LPG injectors turn on, being driven from a standalone ECU. Perhaps i'll track one down and give it a go.

 

[DaveL485]

Good stuff, sounds promising. MOT retest then? I recall scrapping with high and low impedance injectors when the Adaptronic first went on, sacked off the OE ones in the end for similar to what you're using

 

[Ben12345]

Good stuff, sounds promising. MOT retest then? I recall scrapping with high and low impedance injectors when the Adaptronic first went on, sacked off the OE ones in the end for similar to what you're using

No retest needed.... I passed, barely! I just wanted to get the idle CO sorted sooner rather than later otherwise i'd forget until next year. Been using it daily for a week now and loving it. Well apart from the leaking sunroof and this rain that is. And i'm still only on low boost.

Ordered the injector emulator and a dedicated hi Z injector driver.... Will keep everyone posted.

 

[DaveL485]

Well apart from the leaking sunroof

You and every other 21 owner ever (in the UK anyway). French sunroof was optional so a lot of them have full roof skins! Sunroof was factory fit on ALL UK models even the 1.7 poverty spec.

 

[Turbjj]

You and every other 21 owner ever (in the UK anyway). French sunroof was optional so a lot of them have full roof skins! Sunroof was factory fit on ALL UK models even the 1.7 poverty spec.

Surely because there's more Sun and less Rain in UK than in France

 

[Ben12345]

Well I've kind of made some progress, but gone backwards too. Sunroof was leaking where the rubber meets the glass, so squirted some silicone in there and that seems to have done the trick.
Adjusted the drivers door striker, was adjusted way too low before and closing the door was a ball ache.

Took it out for a spin today and noticed a squealing belt. Had this earlier in the week too, just tightened up the alternator, which I did from the top side because I was at work. Got underneath to do it today and saw that the belt that's perhaps 50 miles old is now about a quarter of an inch wide! Hopefully it'll last until Tuesday when I can pick up another. The one I was supplied last time was a make I hadn't heard of, in size 10x1000, which seemed a bit narrow when on the pullys. Ordered an 11.9x1025 this time, and a contitech too. Was the closest length to what the book says without having to wait too long for delivery.
Oh, and while I was underneath, I noticed a drop of coolant coming from the weep hole in the water pump. Great! Should have stayed away from Dolz.

Still not sorted the boost out either, I put in my greddy boost controller, but I think that's packed up too since I last used it, it just doesn't want to calibrate to the car. It's a weird one where you have to set it into learning mode, then do a full throttle 3rd or 4th gear pull, then do it the same again. It completes ok without errors, but I think the valve is knackered because it's still only giving me low boost. I'll have to take it apart and have a look when I've got time.