Chain rattle solution. Easy ignition circuit isolation switch.

[jjrichar]

We own a C5 Allroad with the 4.2L V8. It had the standard chain/cam adjuster rattle.

I initially had a fuel pump cutout switch installed so I could crank the engine with no fuel pressure to build up oil pressure. This worked OK, but if there was some residual fuel pressure then the engine might fire.

I then installed an Accusump, which worked fine, but was some expense and effort to install.
Link:
https://www.audizine.com/forum/showt...st-impressions

About a year and a half ago I did the chain service. All went fine, but I always wanted to stop a repeat of the problem, so I re-installed the Accusump.

Over the last couple of months I started looking through the wiring diagrams to see if there was a solution that worked better and was easier to do. I wanted to find a way of cranking the engine and have it not fire by isolation of either the ignition coils or the fuel injectors.

As it turned out, there is a very easy way of isolating the ignition circuit. If a switch is put across this circuit, you can crank the engine for a few seconds until you think you have built up oil pressure, then close the switch, and the engine will fire immediately. This builds up the pressure in the oil gallery to stop the rattle, but also pre-oils the engine while it’s turning. To me the perfect solution.

Take a look at the applicable portion of the wiring diagram below. Note, this is for a C5 Allroad. The engine in the S4 is identical except for the camshaft and the DME tune. I’m not sure if the wiring is identical. Someone with access to these diagrams might be able to confirm this. Pulling different fuses and turning ignition to position 2 and see what codes appear might also be a way of confirming this.



The green arrows show the direction of current flow. It starts at the top at position 80. This from memory comes from the main DME relay and then travels down to Fuse 229 (fuse 29 on the cabin fuse panel). This then flows on to be the power supply for the ignition coils. D23 is the main supply wire for the ignition coils. N291, N292 are two of the eight ignition coils. All the others are attached to this as well (not shown in my screen shot). For interest, J220 is the DME.

The great thing is that the only thing connected to fuse 29 is the ignition coil circuit. There are 4 different fuses in the cabin panel that are for engine management. The problem is that all the others are for multiple systems. If you isolate the other circuits you can get a similar result as what I’ve achieved, but you will get multiple engine codes and possibly other unwanted results. By isolating the ignition coil circuit you don’t get any engine codes, and it starts perfectly.

If you look at the accusump thread (link above), there are a few pictures showing the fuse panel opened up. It’s easy to also unbolt the relay panel that you can see in the photos, as well as disconnect the plastic wiring guard that is behind the fuse panel which holds all the wires going to the fuse panel. Then you can have easy access to the whole wiring loom.

Make sure you cut into the correct wire. One side of fuse 29 has two wires attached (see wiring diagram above). The other side of the fuse has only a single wire. This is the one you want to cut into. Also, you can see the gauge of the wire is 2.5mm. You need to use a wire to the switch that is similar, and can handle current in excess of the 30A fuse. The switch is mounted to the left of the steering wheel so I can turn the key with my right hand and switch the ignition on with my left. If you are worried about accidentally knocking the switch off while driving (the engine will stop immediately), you may want to use a guarded switch.

If you look at the accusump thread, you will see the wire and switch I used for the fuel pump. I used the same wire and switch, but it now can open the ignition circuit rather than the fuel pump circuit. It's position is such that I would never accidentally turn it off so I haven't bothered with a guarded switch.

So the end result is this.

Accusump removed. Switch installed for the ignition. I simply crank the engine for an appropriate amount of time to build oil pressure (with the switch in the off position), and then turn on the ignition circuit switch while I’m cranking. The engine fires immediately. No fault codes or anything to say it’s been anything but a normal start.

Depending on how long the car has been sitting will determine how long I crank for. The car is a spare that might only be started once a month. If this is the case, I might crank for about 5 secs prior to turning the ignition on. If it has been recently running, I might give it two seconds. Experience will tell you what your engine needs.

Regardless, the engine starts perfectly, like it was running only moments before. You realise why many racing engines are started in this way (crank for oil pressure prior to firing). You only have to listen to the engine to realise it likes it.

Total cost: Less than $10 and about two hours.

[tjbradney]

I didnt read all this but ever replace oil check valves?

[VinnysS4]

Very nice write up. I'm curious why you did away with the acusump system if you already have it?

Sent from my SM-N920V using Audizine mobile app

[jjrichar]

Very nice write up. I'm curious why you did away with the acusump system if you already have it?

I always would have preferred to pre-oil the system while cranking it. It was always obvious from the sounds coming from the engine that it was very good for it. Also, whilst the accusump never failed on me, I was always worried about external oil lines and the possibility of a hose popping off and draining the sump.

To me this is less prone to failure, quicker to start, and better for the engine.

[VinnysS4]

Gotcha

I call dibs on the acusump. Lol

Sent from my SM-N920V using Tapatalk

[emilio]

i don't get why this works, or even if it does anything at all. engine cranking moves the pistons and timing system, so - if oil pressure is the issue - that timing system will be moving with slack in it no matter what. cutting off fuel might spare it full engine power for a split-second, but that doesn't seem like a big difference (and maybe it takes a split-second longer to reach pressure without full engine power). it sounds like your chain service was the big change here.

also, wouldn't it be easier to just floor the pedal to cut off the fuel pump?

- emilio

[dammad]

I always would have preferred to pre-oil the system while cranking it. It was always obvious from the sounds coming from the engine that it was very good for it. Also, whilst the accusump never failed on me, I was always worried about external oil lines and the possibility of a hose popping off and draining the sump.

To me this is less prone to failure, quicker to start, and better for the engine.

I fitted an Accusump about three years ago and used high pressure products without any issues at all, I put a joint to enable front end removal, no leaks. https://daviem.wordpress.com/jh-motorsport/

Sent from my BBD100-1 using Tapatalk

[justinperkins]

engine cranking moves the pistons and timing system

At significantly less RPM, which is key.

[jjrichar]

When you crank the engine, it turns the oil pump and pushes oil through the oil gallery. With a cold engine, the pressure is surprisingly high due to the viscosity of the oil. On a normal cold engine you will probably get close to full oil pressure. The viscosity makes a huge difference. At zero Celsius, the kinematic viscosity of a good quality synthetic oil is about 30 times what it is at operating temp (100 C). At 40 C, it's about 5 times. That means at engine crank speed (about 100 RPM) you will get oil pressure coming out of the pump prior to the pressure regulator that is about the same as 3000 RPM at normal operating temp (if oil is 0 C). Sounds crazy, but this is actually what happens.

So while the engine is being cranked, it's being oiled like it's at idle. The chain tensioners and cam adjusters are being pressurised. As well, all the other moving parts are being oiled prior to ignition. From my reasonably extensive experience with this engine, it seems to me that it's the things break due to rapid acceleration of engine speed at first ignition without chains being tensioned correctly or cam adjusters full of pressurised oil. Having these pressurised prior to ignition to me is the key to stopping things breaking.

[brianup]

When you crank the engine, it turns the oil pump and pushes oil through the oil gallery. With a cold engine, the pressure is surprisingly high due to the viscosity of the oil. On a normal cold engine you will probably get close to full oil pressure. The viscosity makes a huge difference. At zero Celsius, the kinematic viscosity of a good quality synthetic oil is about 30 times what it is at operating temp (100 C). At 40 C, it's about 5 times. That means at engine crank speed (about 100 RPM) you will get oil pressure coming out of the pump prior to the pressure regulator that is about the same as 3000 RPM at normal operating temp (if oil is 0 C). Sounds crazy, but this is actually what happens.

So while the engine is being cranked, it's being oiled like it's at idle. The chain tensioners and cam adjusters are being pressurised. As well, all the other moving parts are being oiled prior to ignition. From my reasonably extensive experience with this engine, it seems to me that it's the things break due to rapid acceleration of engine speed at first ignition without chains being tensioned correctly or cam adjusters full of pressurised oil. Having these pressurised prior to ignition to me is the key to stopping things breaking.

I’ve also cranked the engine without letting it fire to try and build oil pressure. I’ve been doing this for about 50k miles. It eliminates my chain rattle. If I don’t do this, I get a little rattle on every cold start.

I like the OPs idea here. However, I can build the oil pressure my car needs to eliminate chain rattle manually with no mods. I did just replace my starter though... :)

[emilio]

wow, that is strange. maybe the pressure regulator has a variable response so the pump can run unimpeded at startup to reach operating pressure faster. still, i wonder if it is purely the number of pump cycles rather than the speed those cycles are moving.

i've considered the Accusump route, which seems like the best option by far - except for the installation and risks. instead, i've been curious about using an auxiliary electric oil "scavenge" pump: it would need less room and have shorter oil hoses, but it might take longer to reach pressure (assuming there's an affordable one with enough pressure).

- emilio

[jpmiceli]

i don't get why this works, or even if it does anything at all. engine cranking moves the pistons and timing system, so - if oil pressure is the issue - that timing system will be moving with slack in it no matter what. cutting off fuel might spare it full engine power for a split-second, but that doesn't seem like a big difference (and maybe it takes a split-second longer to reach pressure without full engine power). it sounds like your chain service was the big change here.

also, wouldn't it be easier to just floor the pedal to cut off the fuel pump?

- emilio

I tried cranking with accelerator floored. Car just started real fast


Sent from my iPhone using Audizine

[jjrichar]

I don't know why you would put the accelerator to the floor to start. When you put the ignition to position it turns on the fuel pump for a few seconds to prime the system. It then waits for a minimum RPM (ie the car has started) to turn it on again. The only way to stop this priming of the system is to pull the fuel pump fuse, or put a cut out switch like I use to have.

Here's something I wrote a while ago that might explain a bit better what is happening in the oil pump and regulator at different speeds and temperatures.



An oil pump’s output flow is directly proportional to its RPM. So if the RPM is 1000, and the output flow is 10 litres per minute, at 2000 RPM, the output will be 20 litres per minute. This is irrespective of the oil viscosity.

What does depend on the viscosity is the oil pressure. They are inversely proportional to each other. This means the thicker the oil, the more pressure is required to move it, so the pressure is higher.

A simple way to visualise this is to imagine a squeezy bottle of honey. Put it in the fridge so it is cold and very viscous. Remove it from the fridge, and now squeeze it hard enough so that you can empty the bottle in 10 seconds. The amount of force to do this is lots, meaning the pressure needed to move the honey at this viscosity is very high. Now put it in the microwave and heat it up so it is really liquid. Now squeeze the bottle hard enough so that it empties the bottle in the same 10 seconds. The amount of force required to do this is far less, meaning the pressure required for the same flow is far less.

Let’s consider a particular oil at different temperatures and hence different viscosities.
Temp A (cold). Oil viscosity = 40 Cst: At 1000 RPM. Oil flow = 10 Lpm. Oil pressure = 60 psi.
Temp B (warm). Oil viscosity = 20 Cst: At 1000 RPM. Oil flow = 10 Lpm. Oil pressure = 30 psi.
Temp C (operating temp). Oil viscosity = 10 Cst: At 1000 RPM. Oil flow = 10 Lpm. Oil pressure = 15 psi.

So we can see that the output flow from the pump is constant, but the pressure in the oil lines is determined by its viscosity, which is determined by its temperature.

Now let’s increase the engine speed to 2000 RPM, which will create a flow of 20 Lpm.
Temp A: Pressure = 120 psi
Temp B: Pressure = 60 psi
Temp C: Pressure = 30 psi

This is all fine, but the output from the pump has a pressure regulator. This is present so you don’t blow oil lines in the car with high pressures. It’s a very basic item that has a spring loaded piston that compresses when oil pressure is applied. When it reaches the limiting pressure, it allows oil to be returned to the pump rather than go into the engine/transmission, and hence the pressure never gets above this limiting value. This limiting value changes for different engines/transmissions, but a number that is close is 60 psi.

Let’s look at the example above. What you will see is the pressure never gets above the limit of 60 psi, and hence the flow is limited at this point.


Temp A:
1000 RPM pressure=60 flow=10
2000 RPM pressure=60 flow=10
4000 RPM pressure=60 flow=10

Temp B:
1000 RPM pressure=30 flow=10
2000 RPM pressure=60 flow=20
4000 RPM pressure=60 flow=20

Temp C:
1000 RPM pressure=15 flow=10
2000 RPM pressure=30 flow=20
4000 RPM pressure=60 flow=40

What should be obvious is the warmer your oil, the higher you can rev the engine prior to the limiting pressure being reached. When the limiting pressure is reached, the flow will be much higher. Clearly this is good for the engine. Greater flow at higher RPM is a good thing due to the greater load on the engine.

Conversely, what should be clear is that at cold oil temperatures, the limiting oil pressure will be reached at a low RPM, and increasing the RPM creates no more oil flow, even though you are putting far more load on the engine.

The oil temperature numbers I've used here are for pretty warm oil. At zero C, the oil viscosity will be measured in the hundreds of CSt, whilst at normal operating temp it's much lower. Mobil 1 0W-40 for example has the following viscosity specs. 100 C viscosity = 13 CSt. 40 C viscosity = 71 CSt. -40 C viscosity = 21000. They normally don't quote the viscosity at 0C, but this is a temperature where a lot of cars are being started. It's normally in the range of 300-400 depending on the grade of oil.

Hopefully from this you can see why you can actually get quite high oil pressures (even if the flow is low) at very low RPM if the oil temperature is low.

[VinnysS4]

I don't know why you would put the accelerator to the floor to start. .

A lot of cars have a "clear flood crank " which is when you put the accelerator pedal all the way to the floor, crank the car and it cuts the fuel . Great for doing relative compression tests and other diagnosis that requires a crank and no start.

Sent from my SM-N920V using Tapatalk

[jjrichar]

A lot of cars have a "clear flood crank " which is when you put the accelerator pedal all the way to the floor, crank the car and it cuts the fuel . Great for doing relative compression tests and other diagnosis that requires a crank and no start.

I have learned something new. Thanks.

[stevemiller333]

Will this cause the starter to fail?

[VinnysS4]

Will this cause the starter to fail?

It will use the starter more than if you didnt use this method, thus shortening its life (probably an extremely small amount), but it won't make it fail.

Sent from my SM-N920V using Tapatalk

[stevemiller333]

I am going to do this, thanks for the detailed post.

Ordering this switch, probably not in red though, well.. maybe red.

[jjrichar]

Be careful with switch selection. The fuse is 30A. The switch needs to be able to handle at least 30A. The fuse must be the weakest link. The switch above is only 15A.

[VinnysS4]

Be careful with switch selection. The fuse is 30A. The switch needs to be able to handle at least 30A. The fuse must be the weakest link. The switch above is only 15A.

Why not just use a relay? That way you can use any ol switch...

Sent from my SM-N920V using Tapatalk

[stevemiller333]

Be careful with switch selection. The fuse is 30A. The switch needs to be able to handle at least 30A. The fuse must be the weakest link. The switch above is only 15A.

Thanks, will do.

[jjrichar]

Why not just use a relay? That way you can use any ol switch...

I actually think this is a better idea. It would be a bit more work, and you would need to put a fuse in line with the switch circuit, but there are plenty of terminal 30 points available to use near the fuse that would work perfectly.

[VinnysS4]

This gives me an idea. Since I did a manual swap on my car I still have my automatic steering wheel with the paddle shifters. I wonder if I could wire up a relay to the paddle shifter and use that to start the car

Sent from my SM-N920V using Tapatalk

[jjrichar]

Regards using a relay rather than a direct switch. After thinking about this for a bit, unless you can find a place to tap off terminal 15 rather than 30 (ie ignition needs to be on to made the relay to work rather than permanently from the battery) you would need to ensure you turn off the switch so there isn't a constant battery drain. That being said, turning the ignition off is a perfectly good way of stopping the engine when you turn it off.

Regards the paddle shifter. Most paddle switches are a simple micro switch that gives a momentary close of circuit signal to the TCU. You would need to use some sort of switch that you can either turn on or turn off. You also want to have the switch in a place where you can't accidentally knock it off. It could really spoil your day.

[roboto_1337]

Finally a practical use for the "FUNK" button?

[Jdsb6s4]

This idea looks like its going to save me some time and headache.. Probable get this done before my first track day Saturday... So to recap just a toggle switch that can handle 30a or higher and the correct gauge wiring (16 gauge) and we are good to go?

[ArthurPower]

This is a brilliant idea! It also doubles as a kill switch to prevent family members from driving your car. :)

[Velox]

Excuse my noob question, I bought a 30 amp switch today but it has a small led light and 3 connections. 1 is ground I suppose? where do I connect this the best without my car exploding?

[dammad]

This is ridiculous, you can replace the two none return valves very easily or do the job properly and fit an Accusump, any chain slop will damage the plastic chain tensioners and almost certainly lead to the end, ie, if your unluckily enough for the timing to go out of sync, that's it.

Excuse my noob question, I bought a 30 amp switch today but it has a small led light and 3 connections. 1 is ground I suppose? where do I connect this the best without my car exploding?

Sent from my BBA100-2 using Tapatalk

[VinnysS4]

If the switch has an LED light and 3 connector posts, one is for a ground for the LED light and the other 2 are for the switch to work properly.

Excuse my noob question, I bought a 30 amp switch today but it has a small led light and 3 connections. 1 is ground I suppose? where do I connect this the best without my car exploding?

Sent from my SM-N920V using Tapatalk

[romanrnd]

Is this something somebody with absolutely no wiring experience and basic mechanical experience (brakes, plugs..) can do? Lol

[Djscottyc]

I may have to give this a go sometime. I’m a sucker for extra switches.

[Mick288]

Does anybody have a picture of where they mounted the switch? I would like to do this but also want the keep the car very OEM looking.

[Ottawa RS4]

Evening jjrichar, everyone

I had my mechanic install this in my 2007 B7 S4 (same issue as a number of forum members - chains and guides changed, oil check valve replaced and still mild rattle at start-up). The challenge I'm having is that even if I come home and park the car, turning on the switch so the car stalls out at idle, upon start up the next day cranking over, the car stars for a second or two as there is some fuel remaining in the system.

Any thoughts ? is it wired incorrectly ? Any feedback would be appreciated.

Thank you

[VinnysS4]

Evening jjrichar, everyone

I had my mechanic install this in my 2007 B7 S4 (same issue as a number of forum members - chains and guides changed, oil check valve replaced and still mild rattle at start-up). The challenge I'm having is that even if I come home and park the car, turning on the switch so the car stalls out at idle, upon start up the next day cranking over, the car stars for a second or two as there is some fuel remaining in the system.

Any thoughts ? is it wired incorrectly ? Any feedback would be appreciated.

Thank you

Do you have an accusump or just an ignition kill switch?

Sent from my SM-N920V using Tapatalk

[Ottawa RS4]

Just ignition kill switch - I provided the diagram and wiring from jjrichar's diagram to my mechanic.

[1975audi]

Vinnys4
Have you tried this?


Sent from my SM-G955W using Audizine mobile app

[idotavant]

I am at 140k miles. No service. No rattle. I think* I might hear it every once and a while, but it’s for maaaybe .5 sec? Exhaust is loud and the first fire of the pistons covers any other sound, but it’s very quick, and I really don’t hear anything definitive

That being said I want to keep it that way. I’ll be doing the check valve, anyway, but it might be moot with this switch added

Have there been any issues? Updates? Is it a sure way to keep it from firing? I’ve pulled the fuel pump fuse, and it worked, but it hadn’t been started in days before that. I do that when I change the oil

I like this idea, and I feel it would be a great solution for most IC engines, as cold start up is supposedly the bane of their existence. I wonder, though, if oil pressure and it’s lack are so critical/destructive; and cars could be easily designed to crank with delayed start; why aren’t they?


Sent from my iPhone using Tapatalk

[DSchu85]

I was wondering if anyone had a solution for this regarding a B7 S4?
There is no fuse #29 used in the driver’s side fusebox. I’m just wondering where else I can tap into the ignition coil circuit?
Any help is MUCH appreciated

[Rol-co]

I was wondering if anyone had a solution for this regarding a B7 S4?
There is no fuse #29 used in the driver’s side fusebox. I’m just wondering where else I can tap into the ignition coil circuit?
Any help is MUCH appreciated

It is 28, most left one third row from bottom up 20 amps yellow.
Look at the cover you took off.