Mercedes SLK World banner
Status
Not open for further replies.

Preparing for performance mods pre fl 230. CAI

12K views 28 replies 8 participants last post by  Tor Rabe 
#1 ·
Background
So I figured when starting the performance modding of my -99 230, I would just start with the start - the cold air intake, air filter box etc pre supercharger. So to know where the bottlenecks are, I made up a DMDPG inspired from these articles Eliminating negative boost. Here is my DMDPG. :



I made three points of measure. 1st just in front of the intake funnel, behind the grille:



The next point of logging data is in the air filter box, below the filter, hence the difference between no one and no two would give the pressure fall trough the intake snorkel.
The third point was the hole in which the oil catch return hose enters the air filter box AFTER the filter, hence difference between no 2 and no 3 would give the pressure fall trough the filter.
Ideally, I would have measured just at the supercharger intake, to see how much is lost from the box to the SS, but I was not able to make a convenient connection here, so that part of the system is still without data.

I made a connection point where I could swap the three hoses and lead them into the cabin:


Then the hose was lead under the hood, around the A-pillar and in trough the window:


Audun, my assistant, ready to note negative boosts:



Method
Pressures was measured at WOT in 2nd gear (MT) in the same piece of road, slightly uphill, from 2000 - 6000 RPM, and peak value (typically at the high rev end) was noted as accurately as possible

Results
Measure point one, behind grille:
4,5 inches of water negative pressure
(Interesting part is, at 100km/h, at cruising conditions, there were 2 inches of positive pressure here)

Measure point two, in air filter box, before air filter:
17 inches of water negative pressure

Measure point three, in air filter box, after air filter:
18 inches of water negative pressure

Discussion
While the data shows that the SLK cold air intake and air filter system is probably no worse than that of the Audi S4 and Subaru Liberty RS, there is a lot of parasittic drag in the system, of which the filter comprises only 5,5%, the pre intake duct area 25% and the intake snorkel a massive 69,5%!
Needless to say, there is a new snorkel design on it's way. Also, with a better system for collecting air in to the snorkel, there is a potential of getting rid of another 5-6,5 inches of water of pressure fall, converting it in to a positive pressure exceeding the drop over the filter given a good design.

Comments are welcome

cheers,
Tor
 

Attachments

See less See more
10
#3 ·
Lots of work, but good info. Having worked with textile air filtration and waste collection, I learned that the smallest opening is what is going to determine how much flow you have. If you look at the snorkel, the intake opening is about 2/3s the size of the connection to the air filter box. This is all the air you will ever get into the engine. I removed the complete snorkel assembly and replaced it with a 2-1/2" to 3" connector on the air box and ran a piece of 3" flexible intake pipe to where the snorkel originally was. I don't have any hard data, but I'm sure it flows a lot better than it did. One thing you may try to find that would make your testing easier would be a Magnehelic gauge that is used for measuring vacuum and pressure. Dwyer Differential Pressure) They come in many range scales and would make your setup and testing a lot easier. No more tubes and water columns in the car! Very interesting project and I, too am interested in your results. Keep us posted!
 
#4 ·
.... If you look at the snorkel, the intake opening is about 2/3s the size of the connection to the air filter box. .....
Thanks for the interest in my project! Actually, the start of the snorkel is slightly less than half of the connection to the box, roughly 18 sq cm vs 38 sq cm. And the intake is even smaller. This is a pre fl though, I don't know whether the design was made less bad with the face lift model. I'll post more of the progress shortly

cheers,
Tor
 

Attachments

#6 ·
I do not know, and it remains to see wether it has any practical effect on a super charged engine as well. But logics say that it is more efficient if the engine, or the supercharger, do not need to work against a negative pressure. 17 inches of water negative pressure equals 0,6 psi, and I think the 230K engine has a pressure of 7 psi in the manifold. One would expect higher pressure in to the supercharger would produce higher pressure out of the supercharger.

Cheers,
Tor
 
#9 · (Edited)
Post testing CAI

This turned out to be somewhat disappointing, mostly because of the limited data from the pre testing. These are the hard facts:
Original CAI Custom CAI v.0.1
Inches of water Behind grille 4,5 1,5
negative pressure
@ WOT / 6000 rpm Airbox pre filter 17 14
Airbox post filter 18 not tested
However, it was clear to us that there were a significant difference at lower RPM's, stil WOT. The pre test had high values over most of the speed range, where as th post test had mostly very low values, almost up to red line.

So I decided it was desireable to get rid of the old snorkel and went on with a new CF design



Then I got rid of all the gitters and tea strainers in the intake duct,



refurbished the calipers and put on new rotors, ceramic pads and Vmax stainless steel braided hoses

 

Attachments

#10 ·
Supercharger modifications

Next up was removing the h-bend in the air box returning warm air into the supercharger, smoothing and polishing the rough interior of the intake snorkel for the SS and porting the SS. Then replaced the bearings and the oil.


I put on an oversize belt pulley and machined down the SS pulley to 87 mm, increasing the gear ratio from 1,97 to 2,55, giving the SS 15300 spins a minute at red line

 

Attachments

#11 ·
Exhaust modifications

I had a leak in my exhaust manifold, a crack in the welding at cyl 1. This was a perfect excuse for ordering a supersprint 4-2-1 stainless steel header.



At the same time, I ordered the Supersprint muffler



However, Supersprint insist on using 2" pipes, and they also managed to put in a lot of unnecessay bends pre the muffler, so I had to modify them a bit, making it all 2,5" with a race cat in between.



I have an adjustable 3-5 bar fuel pressure regulator lying around that will go in along with the full remap, hopefully quite soon. In the mean time the car runs with erratic behaviour between 3000-4000 rpms, but otherwise very promising. The sound is horrifyingly extatic but unfortunately, there is a bit too much of it, so a free flow slim muffler is in the mail.

To be continued

Cheers,
Tor
 

Attachments

#14 ·
It goes into limp mode after MAF reaches 4.85 volts. (Verified by Kleeman somewhere) I have both 4.7 and 4.3 volt zeners. I now recommend the 4.3's. Reason is they are usually +/- 5% so all the 4.7's I ordered were duds. All measuring 4.87 to 4.98. Even some 4.3's were measuring 4.8x... so I settled with one at 4.75.
You simply strip some wire insulation from pins 3&4 (preFL only!!) staggered so exposed wires don't touch. When wrapping the diode ends, the side with the stripe goes to the yellow/(brown?)pin4 and the other to the brown (pin 3). Then measure between pins 4 & 5 with a multimeter for voltage below 4.85.
A 1W 4.3v zener diode is what you need. A ten pack is like $3 on eBay if none can be found locally.
 
#15 ·
Thanks. Question, Any idea what I would need for a boost of around 19 to 20 lbs? And is this all I would need to pass smog checks? I can build the mechanics to take that punishment but the electronics scares me. I'm shooting for 300+ Hp. I thought of a V8 (to much of a problem) or a 45AMG 2 ltr (too expensive) and both of them would have an electronics nightmare.
Appreciate your help.
Dave F.
 
#16 ·
I don't want to burst your bubble... but that high psi is not possible with our m62 supercharger. 12-14 is max. Even then efficiency is horrible. Lots of heat created. The only way to go higher is turbo... not easy. So... fully modded your looking at maybe 260hp max lol.
 
#17 ·
Mapping / Dyno results

So I finally got my baby on a dyno for a full remap.
First, I was somehow disappointed about the boost, ended at a max around 0.88 Bar (12,7 PSI). I was really hoping for 14 PSI.

Next, as predicted by 1978L82, we encountered problems with the MAF. I had the 4.7V and 4.3V Zeners ready, but using these avoided the P0100 and the limp mode, but instead led to an instant closing of the throttle. We ended up using a Dastec Module to clamp the voltage, which works fine but added another few bucks to the budget. The fuel pressure was set at 4.4 Bar static.

The guy doing the mapping is a recognised expert on Bosch ECU's and I am very impressed with how he has made the engine run. It is smooth, no low areas whatsoever, the redline was lifted 400 rpm's and it is full of evenly distributed power from 2000 - 6000, and feels like a rocket.

The numbers, however, was not that impressing. He was using a Dyno that gives more dynamic resistance, hence it is much better to use to achieve good drivability, Roger's words. (and driveability there is!) But this dyno is cheap on HP readouts, he said his other dyno wold have given at least 10 BHP more. The numbers were 243,1 engine BHP @ 5130, max torque 335,4Nm @5090. However, he was never doing a power registration run, so I guess the max torque might be a bit better at a lower rpm, wot only above 5000 tested. Pre mapping numbers were 205,6 BHP @5545 / 271,6 Nm @5135, same testing conditions (orig spec says max torque 280Nm @ 2500!!rpm)

Discussion: Why was the boost so much lower than expected? I'm pretty sure there were no leaks. The intercooler seems to do the job, highest recorded IAT during two days of testing was 48 deg C, room temp 27 deg C. It seems my supercharger belt is quite slack, it is definately at least 1cm out of it's orbit at speed seen from the paint disappearing from a near lying rotor. My belt is for a larger engine pulley, but not corrected for the smaller supercharger pulley. A shorter belt is in the mail at the moment. Hopefully it will spin the supercharger a bit faster. The next thing to explore is probably the pre supercharger tubing, however, I would like everything to look untouched, so I am hesitating to do anything here. I did a new carbon fibre pre airfilter box snorkel however, that made a huge difference in loss at middle rpm, not so much at red line... However, I'm quite sure, from measurements and from the design in areas I have not been able to test, that there are quite considerable amounts of parasittic drag before the air enters the supercharger.

Comments always welcome,
cheers,
Tor
 
#18 ·
I am also stuck at about 12. Reasons for low boost include leaks, worn supercharger coating, worn belt tensioner, and for some reason if the cam adjust magnet is not functioning properly. I think a new tensioner would help more than a smaller belt. I have belt slip issues as well. I also need to replace my magnet as it leaks. If you want to leave the air box intact, what I did was get rid of all the "h" tubing from the lid and cut a hold on the side. It becomes substantially louder, but I like the banshee whine
. I'm working on a supercharger bypass switch so that it's not so loud on cold starts.
 
#19 ·
I am also stuck at about 12. Reasons for low boost include leaks, worn supercharger coating, worn belt tensioner, and for some reason if the cam adjust magnet is not functioning properly. I think a new tensioner would help more than a smaller belt. I have belt slip issues as well. I also need to replace my magnet as it leaks. If you want to leave the air box intact, what I did was get rid of all the "h" tubing from the lid and cut a hold on the side. It becomes substantially louder, but I like the banshee whine lol. I'm working on a supercharger bypass switch so that it's not so loud on cold starts.
I appreciate your comments!
Well, the new belt is on. It feels snugger, I suspect the tensioner was at it's limit with the other belt. Need to get a gauge to check the boost.
I do not understand what you mean with supercharger coating?
And how would a tensioner wear, except for the bearings?
The air box lit is completely stripped inside, and I cut the plastic where the tube is leading to the supercharger as well. Love the sound;-)

There were actually two moore issues after the remap. First, the engine felt very erratic at idle, after a few days I understood that it was directly connected to the operation of the brake pedal:surprise:. So finally, I got around to measure the supply for the Dastec module, and yes, the +12V disappears when operating the brakes. So I need to find another +12V supply for the module, think I will take it from the +12V to the MAF, from K40 pin A2. Thoughts?
The other issue is a P0100 beeing thrown at me from time to time. No symptoms. Thought it was connected to the breaking issue at first, then I thought not, because it only appeared after a run to +5500 rpm. But then again, I might have been hitting the brakes after those runs, and that might be the connection after all....
 
#20 ·
The coating is the teflon coating on the supercharger rotors. They should be gray. The MAF only gets 5v so you may need another source. And have you re-gapped your plugs to 0.8? Also, I find that the a/c on these cars either uses a lot of power or causes the ECM to pull timing, because vacuum at idle goes from 20-15in/Hg to 10in/Hg. Oh, and p0100 is a MAF sensor problem. Try cleaning it. If it stays the same, replace it. Mine would cut off my supercharger at high rpm (no fun lol) til i restated the car. Then I had a fluctuating idle til I just replaced it. Got a used one on eBay. Google the Bosch number on the sensor. It was used in way more vehicles than ours or even Mercedes. And it's a lot cheaper for other brands for some reason.
 
#21 ·
Thanks! My coating is fine.
The MAF get a 5V reference voltage on pin 4, and a supply voltage of ~14V on pin 2. I used the supply voltage (from pin A2 on the K40) to supply my Dastec module, works perfect now.
I have no current problems with my ignition. I'm using 98 octane. If I would ever experience some problems, I would change from the current 3-electrode NGK to Bosch with similar tempindication (a bit colder) and single electrode and reduce the gap to .6 - .8, but as it is now, it's working just fine.
The P0100 was introduced during mapping (inducing limp-mode with supercharger cut off above 5000 rpm) and caused by the MAF giving a signal above 4,85V, which was the reason for the introduction of the Dastec module. Now, with the Dastec on a steady power supply, the problem is solved. In hindsight, maybe a swap to a MAF insert with slightly different characteristics pre mapping would be the way to go.
When I first got the car, I had several symptoms including radiatorfan allways running, limp mode and P0110? which turned out to be caused by an identical looking MAF insert with slightly different # and different resistance in the temp sensor being mounted by the dealer (suiting a B160CDI or something better than a 230 K).

I got myself a boost gauge today, now needs to find a way to route the hose and go looking for some more boost! (Or removing some more parasittic drag)

cheers,
Tor
 
#22 ·
So what I did this far is (from the top of my head at the moment):
New CAI, the original one is a real power thief (tread on the other forum)
220 mm crank shaft pulley
SC pulley machined to 87mm
Ported M62, with bearings, oil etc per Subby´s advice
Adjustable FPR (set to 4,45 Bar)
Deletion of all plastic/rubber/tea strainers/etc in the intake air plumbing
Remapping

The mapping took me from some 202 calculated DIN-adjusted engine HP to 243, on a dyno the owner said is known to be very cheap on the HP (his other would have given me at least 10 more, but this one was better to work with to get the mapping good, something about dynamic breaking..). I was a bit disappointed with the numbers, but the car runs very well. One of the problems we ran into during mapping was a P0100 and limp mode, caused by the signal from the MAF exceeding 4,85V. Solution was a Dastec module to clamp the max voltage.

However, during mapping, the max boost recorded was somewhere around 12,7 PSI. This was lower than expected. I had not gone to the minor trouble of getting a boost gauge in advance, just supposed the boost would be 14 PSI/1bar. I´m running a 2,53 and redline at 6200, max rev on the SC is 15680. Manual transmission by the way.

This afternoon I finally got around to route a tube to the cockpit, and hooked up a cheap boost gauge (Would like the car to LOOK original/untouched, this was a major step to me). Temporarily placed the gauge in the room next to the ash tray. Took my son for a ride to record boost out of the SC and in the manifold. Did a few 2.nd gear runs. boosts around 17 PSI out of the SC, and 10,5 in the manifold (measured at the FPR). As I said, this is a really cheap gauge, the numbers are not reliable, but after several runs, the difference is. So I think that 6,5 PSI loss from the SC to the manifold seems to be much. I´ve of course visually inspected and tightened all joints, but a leak tester will be manufactured during the next few days. The other odd thing that happened to us this afternoon, was a P0243 and limp mode being thrown at us twice! This code was stored when I bought the car, but this was the first time it returned. Testing the recirculating flap after I got home gave me no ideas on why this code was thrown.

Q: What´s to be expected in boost loss trough the plumbing? Why suddenly a P0243?

cheers,
Tor
 
#23 ·
Psi drop through an air-air intercooler can be high depending on many factors. But the drop is much less through an air-water. I also believe that since a roots compressor compresses air in the piping after the lobes, reducing the volume of the intake system by reducing the length may be beneficial. Eventually I would like to try an air-water setup that passes from elbow directly across to up pipe into throttle body (basically passing between motor and rad to reduce distance). I'm guessing that's why people with this m62 make more boost when it's mounted right on the intake manifold. Same amount of air in less space = more boost.
 
#24 ·
Made up a leak tester today and after a few initial leaks in the pre SC tubes and a couple of cap pop offs, I was able to pressurize the system to around 0,5 Bar/7PSI and it was then obvious that I had a leak at the lower connection of the MAF housing. No need for a spray bottle as the sound was very obvious.
I was able to make up a new O-ring from a bigger and thicker silicone rubber one, cutting to size and glueing with elastic superglue. Tight fit. Pressurized to 10 PSI, no sounds, pressure fell to about the half in around 5 minutes, I think this is as leak proof as you would need in practice.
Then took the car for a spin, engine ran super smooth, no misfires/limps/anything, boosted to a bit over 0.8 Bar/11,9PSI over several runs.
This was better than last tests, and the engine ran super nice, however there is still a drop of 0.3 Bar/4,4PSI which I suspect is around the double of what would be expected on a general basis, however, I have no data on this car.

I run the original IC, but during dyno the highest recorded IAT was 48 deg C in ambient 29 deg C. It is very rarely more than 20 deg C around here, so I do think I´m ok, but still thinking about opening the fender a bit more, as the air intake to the IC is much smaller than the IC it self.


cheers,
Tor
Still searching for those 14 PSI
 
#28 ·
I was looking back in this tread and realized that there have been quite a bit of development since this post.

1. Bought an aftermarket intercooler. Better cooling, better flow. This resulted in poorer boost numbers.
2. Made an extra 2" CAI to the air filter box, routed from the front bumper close to the fog light, via a aluminium duct under the main light to the front side of the box. At the same time I removed the vane inside the supercharger inlet manifold and polished, to the best of my ability, the inside. This resulted in better boost out of the supercharger, around 1.3 bar / 19 psi. Restored manifold pressure to about 0.8 bar / 12 psi.
3. Had constantly recurring problems with leaks where the MAF housing connects to the tight and very bendy piece of plumbing leading from the intercooler. Seems like the pressure deforms the lower plastic tubing and forces the O-ring out. Ended up building whole new plumbing from the IC to the throttle body intake bend, including a relocation of the MAF. I built this new system with 67mm/21/2" PE plumbing, 45 deg bends cut to length and welded together, to get rid of the 110 deg and 90 deg bends at the bottom, and I replaced our 90 deg upper bend with the corresponding part with 45 deg bend from an CLK. The MAf was placed as the last part before the throttle body intake bend. This increased the airflow from around 670 kg/hour to about 705 kg/hour, about 5%. I did not register any significant change in max boost.
4. I have come to a better understanding of the practical application of the law of ideal gases; PV=nrT, and my conclusion is that there is little point in talking about boost in this situation, it´s the air mass that is the interesting number here.

Cheers,
Tor
Still looking for more boost, ha ha
 
#25 ·
Not knowing much about Superchargers and all but would running a 50mm (or so) R/C model ducted in the air inlet have any beneficial results ? You could tie the speed to the rpms in some way I would think... Just something I have been thinking about for some time.

 
#26 ·
Not knowing much about Superchargers and all but would running a 50mm (or so) R/C model ducted in the air inlet have any beneficial results ? You could tie the speed to the rpms in some way I would think... Just something I have been thinking about for some time.
I don't know much about the fans you are talking about, but I doubt they are up for the needed flow. My car currently flows just over 700kg of air per hour (WOT @5.500 RPM), air is around 1,2 kg per 1000 litres, so that would equal around 585.000 litres of air per hour, or over 160 litres per second....
Cheers,
Tor
 
#29 ·
New focus; Traction

With the engine pretty well sorted out last fall, ending my project at around 260-270 HP/400Nm, my focus was turned to traction. The car turned out to have been equipped with some aftermarket lowering springs of unknown origin, combined with the original dampers. This made for a pretty harsh(??) ride while still maintaining a sluggish feel. Not what you long for, actually. It was, however, not until I took my gal to the track I realized how bad she really behaved, tons of understeer and lurching like a 2CV...

So I ordered the Bilstein B12 set, a matched set of Eibach springs and Bilstein lowering sports shocks. Then I figured if pulling everything apart I could as well upgrade to polyurethane bearings, so I ordered a full set from Strongflex. Then I realized the pre facelift didn't even have a rear sway bar, so I ordered the H&R front and rear sports sway bars. When at it, I decided to sandblast and repaint every single part of the suspension. This, combined with the constant lack of available time for amusements, has led to me not having a car ready to roll now that the snow depth is rapidly diminishing around here. And, if this was not enough, I finally found a solution to my terrifyingly annoying inner wheel spin problem as Quaife recently put some new diffs on the market, one of which actually fits the 170 and was mounted earlier today.

More on this to come soon

http://www.slkworld.com/attachment.php?attachmentid=497962&thumb=1
 

Attachments

Status
Not open for further replies.
You have insufficient privileges to reply here.
Top