Sitruunapatonki.fi

I decided to write this thread in english because information about V6 Activas and their problems is scarcely available.

So, I bough year model 2000 Xantia V6 Activa that had been standing outside for year and a half. I should have known better... but hobby is a hobby, so here we go.

This car has many problems and rust spots, but I will get into them later.

First I started to dig into the most annoying issue, the Activa (and Hydractive II) systems complete failure to work. Only basic hydraulics work on this car, Hydractive is FIRM all the time and Activa is soft (SCCAR ball in use, elecronic valve open all the time).

First study with Lexia revealed the horrible truth: Activa ECU Permanent failure code. Activa ECU also contains all functions of Hydractive II, so there is no separate box for HA. Meaning, no electrovalves at all are in use. The ECU still communicates, but diagnostics show unusual data:


The measured operating voltage is only 1.5 Volts! How can the ECU communicate then? Well, Activa ECU has two circuits with separate feeds, digital circuit that gets its power through fuse 3 in passanger compartment (via ign.key), and electrovalve power circuit that gets it feed via fuse 8 in engine compartment.

First challenge is to check the fuses and their connections.

The Fuses (number 8 in engine comp. and number 3 in pass.comp) were fine, so were their connections even from below the fuse box. So something else is wrong.

How to find the Suspension ECU in this year model? Well, documentation suggest it is inside the front right wing, accessible through wheel arch. But no, it was not there.

Internet (from RHD cars...) suggest it is near the coolant expansion tank. But no, it is not in there too.. There is only ABS Unit.

In this year model, 2000 V6 Activa, the Suspension ECU is located right on side with Motronic engine ecu, in front right engine compartment where that large plastic box is.

All connections looked like brand no = bad news, probably not connector corrosion etc. The wiring loom provides right voltages to Green pins 1,2 and 5, and ground comes through fine via white pin 8, measured from connectors.

Well, maybe the electric supply fails only when S.ECU is connected?

I removed the Activa Suspension ECU and took it to bench. It is naturally sealed with silicon glue, so opening required cutting the glue where possible, heating the seal gently with heat blower, and then prying open the ECU. This is what is looks like opened and dismantled:



It does not look clearly broke, there is some heat signatures visible near transistor block (one unit for all 3 electrovalve transistors in this unit!), but no failed solderings or severe heat signs:



So no clear failure points, unfortunately. Murphy...

And then back to the car! I connected the ECU without casing (naturally isolated carefully by placing it on newspaper near the wiring loom, so no short circuits...) and turned ignition on.

Again, Lexia communicates with ECU but still shows only 1.6 Volts of operating voltage.

Measurement from (now visible) connector pins revealed, that pins have correct ground and correct positive 11.8-11.9 volts on every pin where it should be.

So it is not a wiring loom problem... Bollocks! This means back to the bench with ECU. Now I will try to study where the ECu measures its own internal voltage, and why that measurement only shows 1.6 volts.

The project continues...

My best guess is groundings. If you measured against real ground the result is different what ecu see when it measure against it's own ground point. Try to measure between Ecu's own ground pin and power supply pin, if you haven't allready done so?

Wesa kirjoitti:My best guess is groundings. If you measured against real ground the result is different what ecu see when it measure against it's own ground point. Try to measure between Ecu's own ground pin and power supply pin, if you haven't allready done so?

I measured all that. Grounding was good, with less than 0.2 volts difference between ECU and chassis, and between ECU and battery ground.

Now, under suspicion are the several transistors and condensators on board. Measuring continues...

Oops, I noticed from your earlier post..
Is it same 1,5 V when engine is running or bit more in relation to higher system voltage?
How about power consumption? It might tell something? I can measure from H2 the input current, should be in same hectare when in firm mode.
Otherwise cold spray, therma cam, bending/twisting the board, component measuring, 5 min in 175C Owen..easiest Would be To find Another ecu for testing...

Indeed cold treatment and also thermal camera may show the faulty component. I may give a hand with thermal camera. It shows over 110 000 thermal spots on picture.

After measurements it is clear that one VN05N Driver has failed, that one that shows heat signature:



So it is classic Hydractive ECU problem, Activa ECU uses same components mostly.

Now I must wait for the spare to arrive, that component is so rare that it is unavailable in Tallinn. Will take about a week.

On a side note. After looking into the wiring:


..and description of the Hydractive 2 Ecu functions at

http://buukne.dyndns.org/Hydractive.Cit ... Manual.pdf

I could build a new version of Activa ECU myself, using modern components instead of 25 years old tech found in that original box.

The new version would be slightly bigger than matchbox, and include:

- full 1:1 copy of functionality of original Activa BOX (well, probably 0.99:1 copy...)
- build-in "diode fix" of course, the traditional problem that old electrovalves develop
- bluetooth connection for diagnosis and set-up via mobile (android / ios)
- flashing error codes via the sport mode lamp in facia button

Only challenging algorithm to develop would be steering wheel center point calculation from that optical wheel sensor (it has no center point detection, center has to be averaged over time). But not a big problem, just some extra hours of SW development and testing.

Other functions of the Xantia Activa would not be disturbed, Hydractive module seems independent enough from other electronic boxes in the car.

Now when Activa ECU is waiting for spare parts, time to dig into other problems of the car.

Engine runs really rough, spits and sputters on coolant temperature range 20-85. Only when 90 deg celsius is reached it will settle to "normal" idle. Now knowing this power unit and having had many cars with this (and newer ES9J4S) engine, I know this rough idling is not normal at any temperature. Something is wrong. I assume it does not run on all 6 cylinders on idle. Badly leaking exhaust makes it harder to analyze the noise, however.

Lexia does not show much. Couple of anomalities, ignition timing jumps around at idle as much as 15 degrees (strange!) and lambda sensor gives pretty static voltage on idle even after minutes of idling (it should saw between 0.2-0.8 volts all the time, now it does not).

Well, I had info that the engine had been "tuned" by the swedes ten years ago. How? It seems to be classic Unichip piggybag "tune", basically fooling Engine ECU (Motronic 7) with falsifying sensor data. This engine is pretty well optimized by PSA (to run on 98 octane actually), so I cannot believe that piggybag makes much positive difference. However, it can surely cause the ignition problems and maybe also other probs on this engine.

And naturelly it is welded directly into the wiring loom. Those swedes...


So it is a project in itself to remove that shitbox and repair the wiring loom. Not a nice project at that.

Here is the engine running rough at 60 deg celsius coolant temp:

http://youtu.be/iBK8ctVCy7s

And the same from the exhaust, sorry about the audio quality but with high volume you can hear the irregular exhaust note:

http://youtu.be/p1yzezoVWTs

I reckon it's running at least partly on 5 or even 4 cylinders. Comments?

OK, so let's bypass that shitbox (UniChip piggybag ecu) first and see what that does to bad running.

Rather than starting to cut the extra wires out and soldering originals back together at this motronic connector (see quite good swedish solderings in there),


, I decided to make a bridge to the Unichip connector. Bridge will restore the original functions of the Motronic ecu, bypassing totally the Unichip which can be removed then.

As we are in Estonia, it was not hard to find electronics component shop open at 8PM easter saturday. So I went ahead and bought suitable connector (standard PC power block connector) and made prototype of bridge:



Only three connections (6 pins, 2+2+2) needed to be soldered together. Which ones? That was obvious after looking at the new connections and cut wires at motronic connector end.

Plug in and... start...

SUCCESS!

Engine purs smoothly like a cat, cold and warm:

http://youtu.be/LyDIOx4jd04

So it was that Unichip box that was hindering the engine. Now I just cut the extra pins out from bridge connector, and the job is done (after isolating and protecting the bridge from moisture of course).

It is nice to follow people who can and do things!

Sorry about the condination of your project, but now you have enough challenge. More than you asked (or bought), but you seem to like challenges. Anyway, i'm sorry

Now when engine ecu is repaired, and Activa ecu waits for part, I decided to dig into one other annoying problem of the car: deep "thumping" bass noise when driving even on very low speed.

FFT (frequency analysis of this noise) showed that frequency of bass noise was around 20-50 hz and did not change according to vehicle speed, so it was clear it is not from wheels (which was my first guess because modern tyres "square out" when car stands for long time).

After some study and speculation, I found the probable cause. Trunk door feeled little bit loose when it was closed. And behold, there are adjustable rubber studs in the trunk door:



Maybe over time these are compressed, and trunk door can move more than it should, creating this thumping sound?

After adjusting these rubbers tighter and tightening the lock screws to 6 Nm,

SUCCESS!

Bass noise is gone. One point to cross out!

Project continues.

Well, the transistor is now soldered to Activa ECU. Old one was shorted. While at it, I decided to do diode mod also. See those 3 diodes (1 for each transistor) on top of new transistor:


These diodes are common fix, external box doing the same thing is even sold on ebay for 50 euros (the diodes cost less than 1 euro together...):

http://www.ebay.co.uk/itm/XANTIA-ACTIVA ... 2ea5803e7f

There are originally diodes to do this "pull off bounce reduction" inside electrovalves, but they commonly fail. Causing ECU transistors to overheat because they have to take that negative spark every time. New diodes do it now, so transistors should last longer from here on.

Well, all measures OK now! Let's try it on the car:


And power on... attach Lexia... does it connect...

-SUCCESS!

Lexia shows no new error codes after erasing the memorized ECU fault code, voltage is OK, and electrovalves are on soft!



Suprisingly, Lexia does not show actuator test menu for this ECU, so I cannot try out the actual function without test drive. So time to put the ECU back to it's box, install everything correctly back to car, and go for a hard drive!

Project continues...

...and after some soft testing, the Activa system works now correctly:

http://youtu.be/-rdFY_dAIj4

I have some experience about sports cars. For example that Kappa platform Saturn Sky Red Line (Opel GT) that I drive, can corner with about 1G and it is MUCH slower than this old Activa (with winter tyres!).

I think that with right tyres, and normal road surface, this is the fastest car in the world when cornering. And oh yeah, swedes have measured it too, check this our (press hastighet two times to see what the fastest car is in that test...) http://www.teknikensvarld.se/algtest-lista/

Congratulations, great work! Its nice to hear you got it work!
But what makes it the fastest car in corners? Is it only so simple thing that zero body tilting keeps wheels perfectly aligned against road? I don't believe so! Why the real sports cars would not be as fast with almost zero tilting and perfect suspension geometry together with lower COG etc.? It must be somehow related also to juice spring system and it's ability to provide better mechanical grip?

Because suspension is active, Porsche and others cant do this http://youtu.be/kQT7IMHvBGo

I want Activa.

Wesa kirjoitti:Congratulations, great work! Its nice to hear you got it work!
But what makes it the fastest car in corners? Is it only so simple thing that zero body tilting keeps wheels perfectly aligned against road? I don't believe so! Why the real sports cars would not be as fast with almost zero tilting and perfect suspension geometry together with lower COG etc.? It must be somehow related also to juice spring system and it's ability to provide better mechanical grip?

This is a long story. As I have quite much experience in theoretical and practical physics, including friction behaviour, I can speculate. I don't expect all readers to be able to grasp the following, but there are some who will:

So: On normal "road grainy asphalt", why is Xantia Activa faster in transient corners than any supercar, even if supercar itself has also active suspension and "perfect" geometry for cornering (example MB SL55 AMG with ABC active hydraulic suspension, without physical sway bars)?

Well, let's assume Supercar X and Activa are driving through same transient corner, with "normal" rough asphalt. Supercar's center of gravity, or COG, is maybe half a meter lower than Activa - Xantias have quite high stance compared to modern cars.

Both cars are able to pass transient corner with 0 deg tilt sideways, meaning wheels are perfectly flat on the asphalt as suspension is on normal "center" position on all wheels.

But when we look at the weight distribution of the wheels, we notice interesting things. As Activa has much higher COG than Supercar X, it places much more weight during cornering to the outer wheels (in corner radius). Let's say for argument's sake that in 1G cornering speed, 80% of Activa's weight is on outer wheels. On Supercar X, only 65% of weight is on outer wheels.

This makes Activa, on transient moment, "push" outer wheels down to the asphalt with much higher % Newtons in relation to total weight than Supercar X. First of all, during this transient "push" moment, the tyre pushes the road with higher force than the weight of the vehicle normally allows. It can be 110-120% of the force than would be applied in Supercar X, in relation to the mass of the vehicle of course.

So this effect allows Activa to have MORE grip than is should per it's weight, for a transient moment, maybe 0.2-0.5 seconds. And that is enough to go the beginning of the transient direction change faster than any supercar in the world - higher COG actually has advantage here, that may be a big surprise of "low car fanatics"!

Other effect allowing Activa to be so fast is non-linear grip of rubber (tyre surface) to asphalt. As the asphalt is not perfectly smooth, too low weight per square cm can result in effect that tyre "flies" over the asphalt just touching the high points of surface. All modern supercars have very wide tyres and relatively low weight, which makes their tyres easily "fly" over these imperfections in asphalt. Activa, on the other hand, has quite narrow 205/60R15 tyres than maximize the grip on poorer asphalt - specially when cornering the outer wheels really "dig" into the imperfections of asphalt surface, providing extra grip. This can easily account for 10-20% more grip compared to wider tyres.

Last but not least, the Citroen hydraulic suspension provides better following of imperfect road surfaces on corners with rough road. Most supercars are so hard on their suspension, that the body starts to vibrate with road imperfections and dynamically the grip integral is poorer because of the vibrations.

So there you have it. I believe these factors explain why 15 year old Citroen Xantia Activa V6 can pass certain transient corner at 85 km/h, and almost new Porsche 911 GT3 carbon-fibre supercar can pass it at 82 km/h. Audi's best, that many say is the best all-round sports car in the world (Audi R8), can only do that same corner at 80 km/h.

As a last drop, this test and speeds were not measured by "hobbyists" or amateurs, but one of the most recognized technical car magazines in the world, Teknikens Värld in sweden. In fact, this test is so appreciated, that in the end on 90's Mercedes redesigned the whole A-series (and added ESP to it) because original A-series test car rolled and crashed in that Teknikens Värld test!

The effect (2) what I explain is easily tried out with poorer surfaces, snow or gravel, as is commonly known narrow tyres grip better in those conditions too.

"too low weight per square cm can result in effect that tyre "flies" over the asphalt just touching the high points of surface. "

This is one of the real topics about the grip on surface. I have lots of practical experience about this with Citroen GS and 145 mm wide tyres. I havent felt such a grip on any kind of surfaces since I stopped to drive with narrow tyres. Narrow tyre is the key to perfrom safe driving.

Cikatti kirjoitti: Narrow tyre is the key to perfrom safe driving.

I fully agree

Does this engine have any reference about piston flap feature? Or is it only ES9J4S -engines problem?

Project continues.

On the first yearly inspection (regulation in Estonia), there was - well - some surprises.

First the good news: Engine, after I removed that Unichip shitbox, is now running well and passes emissions test with flying colors and good numbers:


Then to the bad news...

Basically the complete suspension system is worn out. Inspectioner found that front axle lower bushes and balljoints are worn out. Also sway bar center supports are worn out. Rear rubber mountings are cracked and worn badly. 3 suspension cylinders "sweat", one has serious leak. Both Activa cylinders sweat.

And fronts brakes are funtioning badly, strangely only exhibited in the slow rolling test. But hydraulic Citroens can do just that when there is small bubble of air in the braking system. So need to air that too, which is actually quite a project with this model, if one wants complete air-free brakes.

Regarding rust spots, it passes inspection still, but next year probably not. So need to weld new side boxes (outer and inner) to place, which is a major project.

Outside inspection, driving home, I noticed that air condition is not working at all. First warm day that I tried to use it, so this is why I did not notice it before. Lexia does not want to connect to Air condition ECU, so need to traditionally start to measure pressures and so on. Or it can be the common fault, engine middle speed fan function not working -> engine ECU not giving clearance to use Aircon. Study continues!

What can I say... I bought a classic lemon... but all the more interesting to you, dear readers, and there will be MANY MANY things to fix.

The project continues. Air Conditioning works now (filled), but it may be temporary. Pump is quite noisy:


Also note the block heater wiring between pump and oil filter. Not the most elegant solution.

New lower balljoints and bushes, and sway bar link:


And finally, bleeding the brakes after correcting the position of brake pad on front brake. Xantias have certain notch in brake pad that must match brake piston, in this case one pad was incorrectly place. Now it is corrected.

Bleeding in progress (Suspension -> high, procedure LF RF LR RR, while all the time keeping at least 10kg pressure on brake pedal, engine running. If back end does not bleed correctly, apply more pressure to brake pedal with pumping motion, never releasing the pedal):


And so the project continues.