vtl's Alpine White 135i

[vtl]

I have sourced a good condition 2nd hand M3 rack and got it installed on the weekend along with the modified 034 engine mounts.

The benefit to the M3 steering rack is the faster ratio 12.5:1 vs 16:1. This is something you can really feel whenever you drive an M3 or 1M. I feel a lot of corners require a lot of steering lock on the non M cars even though the steering feel is already very nice stock.

The M3 rack is a servotronic rack, which means the power assist is variable depending on the speed of the car. At low speeds, the steering is boosted to reduce steering effort and higher speeds the steering assist is reduced. From the drivers seat comparing to the 1M and 135i, the 1M has very light steering at parking speeds and heavier steering at high speeds. The low speed steering of the 1M i don't particularly like at all, feels overboosted. You can feel the wheel tighten up as you increase speed with some steering lock on, which feels very unnatural. This effect really put me off the retrofit, but I recently found the assistance curve can be tuned by NCSExpert to whatever your preference is. My plan is to wire up the servotronic and code in an assistance curve similar to stock, maybe a tad heavier at higher speeds.

This is a list of the hardware components I purchased:

M3 Steering rack + tie rods and boots
M3 steering rack heat shield
M3 billet steering shaft
OEM Lemforder non M tie rods
New bolts for steering shaft

The non M tie rods were purchased as I had read the M tie rods would be too long. I have already got M3 control arms so this information was wrong. I noted with the non M tie rods they have to be extended quite a bit for a proper alignment. If you are doing the retrofit you can use the M3 tie rods if you already have M3 front control arms.

I sourced the M3 steering shaft off ebay as I saw it for a good price. This is a shiny billet machined part which may be slightly more stiff than stock, resulting in better steering feel. I got this mainly because it looks cool, the stock steering shaft does indeed fit the M3 rack.

New steering shaft bolts were used as they are micro-encapsulated screws (has loctite which gets activated on first use) and they are very inexpensive

The heatshield for the M3 rack is 2 piece stainless which has one bolt on one side and dimpled tab on the otherside. In my experience this headshield can pop off and scrape on the steering shaft if its ever bent or tweaked. The second hand heatshield that came with the rack was very bent and I had to spend quite a bit of time hammering it back into its original shape. I spot welded the tab so it can never come apart, then I polished it up with sandpaper and a wire wheel to make it shiny.

I also took the opportunity to clean up the rack of any dirt grease and oil and looks quite clean.

The installation of a steering rack on these cars is very simple but can be time consuming and annoying. The power steering fluid can make a mess if you are unfortunate enough to spill it. I did it by myself but a 2nd pair of hands will help a lot when raising the rack back into the car. Servotronic wiring has not arrived yet so I will have to install it at a later date.

Initial driving impressions were sketchy. I measured the old threads of the tie rods with a digital cliper and made the new tie rods exactly the same length. This resulted in a LOT of toe in on the M3 rack, the M3 rack is actually shorter than stock. This resulted in tyre marks on my driveway as I reversed out onto the street for the first drive. I then took the car back in and did an eyeballed alignment and made each side equal. I added approximately 8mm of additional length on the tie rods to make it drive normally.

After the eyeballed alignment the car drives pretty straight, steering is only 1 degree out on the highway. The faster ratio I noticed immediately and feels great. The car feels so much more repsonsive. The steering is quite heavy at low speeds but as soon as the car is moving its fine. 3 point turns require a bit more effort but its not quite as bad as a car with zero power assistance. I will install the servotronic harnesses once I receive the parts and make up a harness to wire into the junction box.

Overall very happy with the new rack, highly recommended

[Pete.J]

Awesome build! And good idea about the 1M/M3 steering rack, I've been strongly considering this mod to my car as well.

But upon looking up part no. on Realoem.com, it seems that the cars with the servotronic steering also run a different power steering pump, and power steering reservoir. You can probably get away with using your old reservoir, but I don't know if you can use a non-servotronic pump on servotronic steering rack and achieve similar results!

[vtl]

Originally Posted by PeterY

Awesome build! And good idea about the 1M/M3 steering rack, I've been strongly considering this mod to my car as well.

But upon looking up part no. on Realoem.com, it seems that the cars with the servotronic steering also run a different power steering pump, and power steering reservoir. You can probably get away with using your old reservoir, but I don't know if you can use a non-servotronic pump on servotronic steering rack and achieve similar results!

The power steering pump is different and is more powerful on the M3 rack. With the stock 1M steering curve itll be more heavy than a 1M but I plan on using a custom tuned curve so I could theoretically just run more duty cycle on the solenoid to compensate. I'll probably run similar levels of assist to stock non M racks at low speeds and less assist at higher speeds and make it linear so the steering feels more natural.

[Muppet]

I've had a look at the issue of the different pumps and both have a pressure of 65 BAR but are completely different pumps. Maybe it is the pulley size that is different with a larger pulley creating lower pressure in the 135i.

I have been through the coding options in the JBBF ecu and found there is some adjustment. Mine is set at max assistance (comfort mode). There is a thread re this on this site in the suspension forum where I have posted the options.

The weight now is acceptable. A bit heavy some days and nice others depending on the amount of wheatbix in the morning. At over parking speed the weight is perfect.

If you have found a way of changing the steering curve then I would be very interested.

Finally, you will need to obtain the 2 pins that are used in the junction box as well as the plug into the servotronic transducer as they are a bit specialised.

[vtl]

Originally Posted by Muppet

I've had a look at the issue of the different pumps and both have a pressure of 65 BAR but are completely different pumps. Maybe it is the pulley size that is different with a larger pulley creating lower pressure in the 135i.

I have been through the coding options in the JBBF ecu and found there is some adjustment. Mine is set at max assistance (comfort mode). There is a thread re this on this site in the suspension forum where I have posted the options.

The weight now is acceptable. A bit heavy some days and nice others depending on the amount of wheatbix in the morning. At over parking speed the weight is perfect.

If you have found a way of changing the steering curve then I would be very interested.

Finally, you will need to obtain the 2 pins that are used in the junction box as well as the plug into the servotronic transducer as they are a bit specialised.

With the pump the 1M has the same V belt as the standard 135i N54. Also the power steering pulley is already quite close to the subframe, on older 335is with old engine mounts the pulley can actually contact the subframe. So I don't think its a larger pulley. Maybe the pump internally can flow more fluid at the same pressure.


NCS Dummy can be used to see all the available options and coding data. For example, for COMFORT, there are 2 sets of data, one for speed and one for assistance. See picture below for graph of data

COMFORT_GESCHWINDIGKEIT = Breakpoints for speed
COMFORT_STROM = values for assist

The curve looks quite linear until you hit 5kph then the assist is boosted a lot.

You can simply COMFORT_STROM to set the assist to whatever you want within the limits of the hardware. For me, I will interpolate the stock curve and get rid of the overboosted assist at low speeds. If you are happy with the stock curve but want it a little lighter by 10%, you could try multiplying all the values by 1.1 (assuming increasing values result in a linear increase in steering assist) and saving it as a new custom coding option with NCSDummy. Then recode the unit

I have done a lot of electronic retrofits to this platform and have a healthy amount of stock for MQS pins which are used in the JBE connector. I have ordered the servotronic harness for completeness, but if you wanted to make your own it is the same plug as the one found on the side indicators.

[Muppet]

Well there you go.

I kept looking at the Hex values without figuring out the relationship, should have looked at the Dec values.

I have written a new wert_03 value into the STROM lines and added an extra wert_03 for the speed values the same as wert_01 and 02 in case there is a correlation.

The 0 kph value I set at 250 in the new STROM wert_03 and 5 kph was 190. The weight actually increased. I must be doing something wrong or 200 is the highest value (most assistance).

In any case what I did find was that the reason I found that the steering had increased in effort was that the STROM wert values had reverted from 02 to 01. Must have happened when I was mucking around in the KOMBI module a little while ago and reset everything as I do remember having to get rid of the disclaimer again and a few other coding things. Steering is much better again. The STROM wert _02 value is nice.

[vtl]

Originally Posted by Muppet

Well there you go.

I kept looking at the Hex values without figuring out the relationship, should have looked at the Dec values.

I have written a new wert_03 value into the STROM lines and added an extra wert_03 for the speed values the same as wert_01 and 02 in case there is a correlation.

The 0 kph value I set at 250 in the new STROM wert_03 and 5 kph was 190. The weight actually increased. I must be doing something wrong or 200 is the highest value (most assistance).

In any case what I did find was that the reason I found that the steering had increased in effort was that the STROM wert values had reverted from 02 to 01. Must have happened when I was mucking around in the KOMBI module a little while ago and reset everything as I do remember having to get rid of the disclaimer again and a few other coding things. Steering is much better again. The STROM wert _02 value is nice.

From what I've seen in NCSDummy wert_01 and wert_02 have have same values. Is it different on your version of junction box?

Did you notice between 5kph and 15kph the assistance was still heavier than before with teh wert_03 custom coding? Would not make sense it would increase as 190 is still lower than 200 and higher than the previous value. Maybe 250 is beyond a max value and 200 is close to the limit as you said.

[Muppet]

The junction box module after installation of the $216 Servotronic option is JBBFE3R.C40. My car is an Oct 2010 N55 DCT.

I cannot get a screen shot from the small laptop I use for the various BMW tools but the assistance at 0, 5, 15, 30 kph is Wert_01 200, 145, 135, 125 and Wert_02 200, 180, 153, 148.

The assistance is the same for Sport mode and Comfort mode which is a bit confusing.

With the Wert_03 coding the steering was heavier on all counts, as if the servo mode was non operational. As I said probably something I have done wrong. However when I code the change in NCS Expert does not argue about it.

[vtl]

Was having issues with fuel pickup during track driving over the last 2 years. Had replaced the fuel filter assembly with no luck. Under a half tank I would sometimes get a CEL for low fuel pressure. Also the final corner of Haunted Hills hillclimb im guessing I was getting fuel starvation as I would get massive backfires out of the exhaust.

Never had any issues with straight line running

During a test run up mount dandenong on low fuel I happened to have MHD logging running. I was getting backfires/crackles out the exhaust (sounded glorious and amazing) during cornering out of slow turns. Looking at the logs I was dangerously low on fuel pressure from the LPFP:

https://datazap.me/u/vtl/backfire-ce...11-12-13-23-27

Luckily I had a Stage 2 Fuelit pump ready to go. I noticed that upon removing the pump, the one of the venturi lines have popped off the bucket! It is not held in that securely and I definitely remember plugging it in last time i did the fuel filter.

After replacing the pump, no issues, even did a Winton Circuit Club track day without any fuel issues right down to fuel warning light

[vtl]

Winton Circuit Club onboard:

Best time of 1:42.202, improved a lot with good coaching from dancinsteve

Had O2 sensor issues which will need to be replaced, but luckily was able to go to a lower boost map to continue running and enjoy the rest of the day.

Car was great with DSC off, the M3 diff works a lot better on track when the AD08Rs are fully warmed up and at their peak grip. Would be great to have more front end grip though, wider front tyres with some more camber would really help the car turn better

[K-Bag]

Car sounds great. Getting some instruction is hands down the best way to lower your times. Nice shifting also.

How did the car hold up mechanically?

[vtl]

Originally Posted by K-Bag

Car sounds great. Getting some instruction is hands down the best way to lower your times. Nice shifting also.

How did the car hold up mechanically?

Apart from the O2 sensors the car held up well. I had issues with the turbos and fuel pressure but since they've both been replaced all good.

The IATs, oil and coolant temps skyrocketed, really needs better cooling for continued track use

[K-Bag]

Yeah it looks like 10 minutes is all shes got. Any more and we need radiator, intercooler and oil coolers :-(

[vtl]

This is a datalog of a session at Winton:

https://datazap.me/u/vtl/ppk-track-runs-ii

Temps go to insane levels. As you can see MHD cooling targets basically do nothing, not sure why people said it would lol. I've got a stock radiator and stock oil cooler with the AD Eng thermostat blockoff. I have a CSF radiator waiting to go on the car so next time at Winton itll be a good before/after comparison.

Worth noting that the ambient temp at Winton was around 16 and dry, very ideal conditions yet the cooling system was at its limits. In 10 minutes the oil is already close to 140 degrees C

Intercooler wise, I have a Wagner EVO1 performance which is adequate for street driving, but gets quite toasty on track. Winton you have a lot of slow long corners at high rpms so its not getting much airflow.

[Brule]

10 minutes is pretty good.

The Atm intercooler would keep temps reasonable But oil temps would exceed 120f with a block off plate just going with down the straight.

Should try removing the coolant and running 90% distilled water and 10% coolant should keep temps down better.

[Piets94]

Nice running there Vince. Winton is a fun track.
How did you find the new suspension set-up compared with the B12 kit you had in before?

[T1M]

Originally Posted by Brule

10 minutes is pretty good.

The Atm intercooler would keep temps reasonable But oil temps would exceed 120f with a block off plate just going with down the straight.

Should try removing the coolant and running 90% distilled water and 10% coolant should keep temps down better.

To be fair the OEM radiator did pretty well, water temps only reached 110 and weren't climbing too fast. The real problem is the oil temp, which simply cannot be controlled with the OEM cooler - at a guess with longer sessions it would limp mode the car.

[gmx]

I might be blind but do you have IAT log?
15psi is probably on the limit of a circuit car too, would be interesting to see.

Do you have any adjustment with the waterpump PID? I think MHD has finally made this available, I'm not sure if XDFs have though. Might help with oil temps a little bit (not really sure about 35i non is/1M setup).

[vtl]

Originally Posted by Brule

10 minutes is pretty good.

The Atm intercooler would keep temps reasonable But oil temps would exceed 120f with a block off plate just going with down the straight.

Should try removing the coolant and running 90% distilled water and 10% coolant should keep temps down better.

Currently i'm running around 3L of BMW coolant (total capacity is around 8.5L for manual transmission). As its a daily I'm not too keen on running lower concentrations of coolant, would want to keep the anti corrosion properties. Upgraded CSF will be fitted before the next track day so that should help a lot.

Originally Posted by Piets94

Nice running there Vince. Winton is a fun track.
How did you find the new suspension set-up compared with the B12 kit you had in before?

Honestly on the track and at the level of driving i'm at theres not a massive difference between the 2 suspensions. The KW handles the bumps much better (B12 would be constantly riding the bump stops up front). But on the track where its smooth it doesn't matter as much. The main benefit was the stiffer front springs which keeps the rear wheels from unloading as much and gives more traction. But the KW V2s are still quite soft (35N/mm front 79N/mm rear) so it could be improved.

Originally Posted by gmx

I might be blind but do you have IAT log?
15psi is probably on the limit of a circuit car too, would be interesting to see.

Do you have any adjustment with the waterpump PID? I think MHD has finally made this available, I'm not sure if XDFs have though. Might help with oil temps a little bit (not really sure about 35i non is/1M setup).

Check the datazap link and click IAT to add the IAT to the display graph.

Yes its in the XDF, the water pump cooling target was 80 degrees at WOT so the pump was running at max the whole run. It basically does nothing as the radiator is already at its limits. I have a CSF upgraded radiator that I will fit soon so that should keep the water temps down a lot and bring the oil down a little as well.

[vtl]

Another thing to note is that the track running resulting in hitting the EGT limits of the tune. In the logs I can see the car runs great until the AFRs start going very rich.

In the tune there is a table called
"EGT NOx Set Point (CAT Protection Mode)"

All values are set to 1742f (950c)

When this value is hit the AFRs default to 10.77:1! The problem is that this value is not measured and is a calculated value based a mathematical model based on timing, AFRs and boost etc. With the catless downpipes the calculation will no longer be accurate and this value needs to be increased for sustained running.

To confirm this was the problem I encountered on track, I reduced the values for this table to a very low temperature. Upon flashing the modified map and starting the car it immediately went to 10.77:1, now we know.

[vtl]

Previously I recently did a brake disc and pad refresh and noticed the front dust boots were perished and cracked.

I ordered new dust boots to refresh the calipers. I did not expect it to be such a pain. The seals were very difficult to pick out. Real bitch of a job would not want to do this again.

Ran out of brake fluid to bleed the front brakes too, the brakes work but are a tad spongy. Ordered a new bottle of Motul RBF660 500ml

The front calipers are now quite yellowed from the heat from the recent Winton circuit club day. I have had this in the past, usually takes a couple weeks for the colour to change back to normal. Fingers crossed they don't stay like this.

[K-Bag]

Looks like time for new brakes up front :-)

I went to Alcons, shaved 2kg per side plus bigger disks.