My ECU's role

This is the Forum for all your Citroen Technical Questions, Problems or Advice.

Moderator: RichardW

Post Reply
MikeT
Posts: 4809
Joined: 11 Jun 2007, 16:17
Location: Christchurch, Dorset. UK
My Cars: 2005 C5restyle 1.6HDI 16v 110hp VTR Estate
2008 C5 X7 1.6HDI VTR+ Saloon
x 231

My ECU's role

Post by MikeT »

I'm sure Jim's already explained this to me but it still hasn't sunk in fully. I know I have a TPS and Lift Sensor and ECT sensor and that the ECU exerts timing control from them but how exactly does the ECU alter the timing?

Apart from operating the EGR and cold start device, does the ECU control or monitor any other parts of the engine management system?
User avatar
Xaccers
Posts: 7654
Joined: 07 Feb 2007, 23:46
Location: Milling around Milton Keynes
My Cars:
x 184

Post by Xaccers »

Jim will be able to give you the full details, but in the pump there's an electrovalve which controls the advance of the pump.
1.9TD+ SX Xantia Estate (Cassy) running on 100% veg
1.9TD SX Xantia Hatchback (Jenny) running on 100% veg for sale
Laguna II 2.0dCi Privilege (Monty)

DIY sphere tool
MikeT
Posts: 4809
Joined: 11 Jun 2007, 16:17
Location: Christchurch, Dorset. UK
My Cars: 2005 C5restyle 1.6HDI 16v 110hp VTR Estate
2008 C5 X7 1.6HDI VTR+ Saloon
x 231

Post by MikeT »

Jim makes such an effort to assist everyone I tend not to ask him directly anymore. :oops:

The electrovalbe is in the pump is it? That'll explain why I can't see anything back there. I assume it's prone to failure though I have no idea what the symptoms would be if it did.
User avatar
CitroJim
A very naughty boy
Posts: 49610
Joined: 30 Apr 2005, 23:33
Location: Paggers
My Cars: Bluebell the AX, Polly the C3 Picasso, Pix the Nissan Pixo, Propel the duathlon bike, TCR Pro the road bike and Fuji the TT bike...
x 6176
Contact:

Post by CitroJim »

MikeT wrote:Jim makes such an effort to assist everyone I tend not to ask him directly anymore. :oops:
Don't be daft Mike :lol: Only too happy to.

Give me an hour or so and I'll tell you in detail. Just got back from Somerset and off to pick up Siany in a mo...
Jim

Runner, cyclist, time triallist, duathlete, Citroen AX fan and the CCC Citroenian 'From A to Z' Columnist...
MikeT
Posts: 4809
Joined: 11 Jun 2007, 16:17
Location: Christchurch, Dorset. UK
My Cars: 2005 C5restyle 1.6HDI 16v 110hp VTR Estate
2008 C5 X7 1.6HDI VTR+ Saloon
x 231

Post by MikeT »

Hi Jim, thanks, that's very kind of you as always.
User avatar
CitroJim
A very naughty boy
Posts: 49610
Joined: 30 Apr 2005, 23:33
Location: Paggers
My Cars: Bluebell the AX, Polly the C3 Picasso, Pix the Nissan Pixo, Propel the duathlon bike, TCR Pro the road bike and Fuji the TT bike...
x 6176
Contact:

Post by CitroJim »

Sorry for the delay Mike, got talking and cooking!

Right, the ECU looks after pump timing, EGR, fast idle, glowplugs and the tacho. It also lights up a light to say when it has all gone horribly wrong.

One thing is does not do is fuelling. This is carried out mechanicaly in the pump in the time honoured way using a centrifigal governor and variable injection cut-off. All you do with the throttle pedal is change the preloading on the govenor basically. The preload on the governor is further controlled by the maximum fuel delivery adjustment, the fast idle lever and the boost overfuel device.

Timing: The ECU takes inputs from the TDC sensor, Coolant Temperature Sensor, Needle Lift Sensor, Load Pot (on top of the pump) and Atmospheric Presssure. Sensor inside the ECU.

Using information from the Load Pot, Coolant Temperature Sensor and Atmospheric Pressure Sensor, Needle Lift and TDC sensors, it decides the optimum timing to be applied to the pump for the conditions prevailing. It does this by consulting maps. For instance, when the engne is cold, it needs more advance (cold fuel takes longer to get burning) so the ECU consults its cold map. When hot and under heavy load, it'll have a hot and heavy map.

Timing is set in the pump by varying the hydraulic pressure acting on one end of the timing device piston. The timing device moves the pump rollers relative to the pump shaft to vary the timing. The hydraulic pressure applied is controlled by pulsing an electrovalve open and closed. This valve is below the rotor head and comes out on the brown connector. The load pot is on the blue.

The problem is that the electrovalve is unable to set an absolute timing value and it is adjusted in a feedback loop until it is right. The ECU does this by measuring the engine speed from the TDC sensor and measuring the time delay between a signal coming from the Needle Lift Sensor and TDC Sensor. The needle Fift Sensor gives a "Start of Injection" signal. This occurs before TDC so if the time interval between Start of Injection and TDC is known, along with engine speed, the degree of timing advance can be computed. The timing is then trimmed by the ECU in a feedback loop until it is corrrect. You can see it happening on the Lexia. It gives the desired timing and the actual timing set. Both should be identical if the system is working right.

The ECU controls fast idle and glowplug post-heating (i.e. how long the glowplugs stay on after the engine is started from cold) by reference to the coolant temperature sensor (the green one).

EGR is also applied by the ECU and is effective only when the engine is hot and on very light load. It controls it by pulsing the EGR valve open and closed rapidly. I can hear mine fluttering on the 2.1!

If the ECU cannot set the desired timing, it lights the engine management light. Again, on the Lexia, you can see this happening when say, the Needle Lift Sensor is malfunctioning. Desired timing will be say 8.4degrees and the actual will be zero.

If the ECU is disconnected from the pump, the pump will still work but the timing will be controlled only by hydraulic pressure, related to engine speed. In practice, the pressure will build until the pump is running at full advance with attendant engine knock. I always thought the timing electrovalve controlled pressure to the timing device but as a result of my full pump strip, I discover it's the opposite, the electrovalve controls pressure by controlling the leakage away from the piston, not pressure to it.

The tacho is generally driven from the TDC sensor but if it is disconnected, the tacho might still work due to a signal from the needle lift sensor. Xac and I once proved this to be the case.

Hope that makes sense Mike. Shout if anything is not clear.
Jim

Runner, cyclist, time triallist, duathlete, Citroen AX fan and the CCC Citroenian 'From A to Z' Columnist...
MikeT
Posts: 4809
Joined: 11 Jun 2007, 16:17
Location: Christchurch, Dorset. UK
My Cars: 2005 C5restyle 1.6HDI 16v 110hp VTR Estate
2008 C5 X7 1.6HDI VTR+ Saloon
x 231

Post by MikeT »

citrojim wrote:Hope that makes sense Mike. Shout if anything is not clear.
Crystal, Jim. Thanks.

Do you think it would be possible to control the timing by altering the ECT value for advance and Lift Sensor value for retard?
User avatar
CitroJim
A very naughty boy
Posts: 49610
Joined: 30 Apr 2005, 23:33
Location: Paggers
My Cars: Bluebell the AX, Polly the C3 Picasso, Pix the Nissan Pixo, Propel the duathlon bike, TCR Pro the road bike and Fuji the TT bike...
x 6176
Contact:

Post by CitroJim »

MikeT wrote: Do you think it would be possible to control the timing by altering the ECT value for advance and Lift Sensor value for retard?
Not easily Mike, I think the only way you could fool the ECU into changing timing values is to delay the TDC Sensor signal to fool the ECU into thinking it's more retarded than it is under the circumstances you'd desire more advance. It could be done with a custom "piggy-back" ECU as they used to once make for petrol cars to "do a bit of ECU fooling" but that was normally to increase fuelling rather than change timing.

In theory, if the ECU was dead, it might be possible to knock up a circuit to give a controlled signal to the timing electrovalve. This could also be done with a carefully calibrated orifice in place of the electrovalve and then (again in theory) you could use one of these pumps to replace a fully mechanical pump as that is basically how the automatic advance on a fully mechanical pump works.

Why do you wish to play around with the timing?
Jim

Runner, cyclist, time triallist, duathlete, Citroen AX fan and the CCC Citroenian 'From A to Z' Columnist...
MikeT
Posts: 4809
Joined: 11 Jun 2007, 16:17
Location: Christchurch, Dorset. UK
My Cars: 2005 C5restyle 1.6HDI 16v 110hp VTR Estate
2008 C5 X7 1.6HDI VTR+ Saloon
x 231

Post by MikeT »

citrojim wrote:Why do you wish to play around with the timing?
According to tests, the mechanical pumps timing advance is restricted and if my ECU is mapped the same, there's room for improvement. Not saying I want to change the timing yet as I don't know what the maximum advance is on a semi-mechanical. Did you say your Lexia can tell you?
User avatar
CitroJim
A very naughty boy
Posts: 49610
Joined: 30 Apr 2005, 23:33
Location: Paggers
My Cars: Bluebell the AX, Polly the C3 Picasso, Pix the Nissan Pixo, Propel the duathlon bike, TCR Pro the road bike and Fuji the TT bike...
x 6176
Contact:

Post by CitroJim »

Yes, the Lexia tells you what the ECU has decided for the prevailing conditions and what the pump is set to.

I understand a couple of degrees more advance is useful when running veg. Seems logical as veg (especially when cold) will doubtless have a longer ignition delay compared to proper diesel.

Mike, be aware thet even if you move the position of the pump in the cradle, it'll make not a jot of difference to the timing as the ECU will compensate. Given how it measures set timing, the relative mechanical position of the pump is (within limits) not an issue.

There is a finite limit of advance imposed by the pump mechanicals. I'd say, off had, the maxium swing is no more than about 15 degrees, looking at how far the rotor can move relative to the pump body.

If you want to experiment with timing and don't want to go down the TDC sensor delay route (difficult), I'd consider a swap to a fully mechanical pump. I can do you one in exchange for your existing pump :wink:

With a mechanical pump, you can set the initial advance by rotating the pump in its cradle and also boost the advance even more by operating the cold start advance electrovalve when the engine is hot.
Jim

Runner, cyclist, time triallist, duathlete, Citroen AX fan and the CCC Citroenian 'From A to Z' Columnist...
MikeT
Posts: 4809
Joined: 11 Jun 2007, 16:17
Location: Christchurch, Dorset. UK
My Cars: 2005 C5restyle 1.6HDI 16v 110hp VTR Estate
2008 C5 X7 1.6HDI VTR+ Saloon
x 231

Post by MikeT »

Thanks for the IP offer Jim though I'm happy to stick with my semi-mechanical pump because the timing is electronically controlled :wink: Yes, it could be beneficial to alter the timing. They say the full advance, like the fuelling setup, is quite conservative. I would be most grateful if you could confirm that one day, should the opportunity ever arise. As you say though, it is physically limited even if I did gain electronic control so a bit of metal surgery will still be required if more advance is needed.

Though I'm getting too advanced of myself, what with the cylinder head still sitting in my kitchen :lol: All for the want of two clic clips. :roll:
Post Reply