TheSamba.com :: View topic

Wed Apr 25, 2012 1:25 pm

Alright.....

After some trials and tribulations I have found why my car isn't running the way I expected it to exactly. Its been hellaciously rich. Now, I've been fighting this battle on two fronts. Fuel and timing. This motor had been dyno'd before using 40 dells and it ran happily. I needed to get back to that point, however, I had no reference point to start from. So, I started with timing. The engine was running happily on a standard SVDA and REALLY well on a custom curved SVDA. So I looked up the Mexican SVDA advance curve. I took the RPM section and put it into the spark table then crossed it with the additional advance provided by the vacuum supply. It sounds complicated, but its a very simple thing to do when you look at the two things and see how they would line up. I then blended horizontally between the vac values. Viola....I now have a simulated SVDA. I have eliminated one unknown....I can now select the entire spark table and add or subtract single or multiple degrees and simulate turn the distributor for advance or retard. The problem is, where do you start advance or vacuum advance. Easy, 1 you can set it the way the manual says and set your max "mechanical advance" at 3000 or so RPMs and build your "vacuum advance" around that. Or 2, since my engines prior life had dyno numbers from DRD racing, I knew he tuned the motor for 10deg at idle and let the rest land where it may. So, since I was asking DRD for advice to begin with I started here. Now, some more advanced play time would be to basically custom curve my "simulated SVDA". With what you've learned with this process you can then go even farther and tune beyond what an SVDA was ever capable of. At this point, with some understanding and trial/error, you now have a pile of unlimited virtual springs and vacuum canisters on your workbench and one of the nifty distributor curving machines at your disposal.

So...now that I have a better grasp of how these tables act and react I can tackle the fuel. But where to begin? Lets call up DRD Racing again. My dyno sheet says I was running at a very lean 15.5 at 3000 RPM. Holy moly thats lean. However, I've been running that way for the past year with an accurate CHT gauge. I had no pinging and no wild CHT readings. On top of it the plugs read nice and the exhaust ports were clean when I pulled the exhaust for this install. So I figured...must be alright then. As I talked to DRD he gave me some other very astonishing numbers for target AF ratios that he used in tuning this motor. I said thank you very much and off I went to the CB dash board. I opened the Target AF table and plugged in the numbers he gave me at the appropriate ranges and comfortably blended between these points (theres no blend button so you have to make it up as you go). So now...HOPEFULY...its pretty close to a simulated version of what I had before. Now its running much cleaner, I have had soot dripping out of my exhaust for the past week while trying to tune this.

It still needs some tweaking in the idle area as would a carb setup basically, but I'm narrowing down variables which allows me to modify things one at a time and observe the results.

And for those concerned about det and heat. My daily surface street drive today produced 250deg tops where it was reading 260-280 for the same drive. I have no "whisping" sounds while driving under load and no definite "pining" sounds either.

In my opinion...THIS is where to start. Go with what you know worked and modify from there. It helps you learn while you figure out what each mod does while still being in bounds of what you know works already.

Mon May 07, 2012 10:26 am

Well...got an update here....

I have now switched over to TPS tuning rather than MAP based tuning. Between the cam, engine size, throttle body size, and exhaust design I have come to decide I do not have enough vacuum available to tune this way. The clincher was on the freeway I saw I was running 98-100Kpa or 1Bar at 55-60mph/3000rpm/4th gear. I've got no where to go from there. Cruising is supposed to be in the middle of the table....I was all the way at the right.

Having switched to TPS, its now night and day tuning. I'm no longer spread halfway across my table I get to use the whole thing. Idle is much smoother and easier to control. I've got my idle down to 900 with AF of 14:1 and its pretty happy there. My idle doesn't jump around as the vacuum changes...which can be compensated for, but it makes it a little more complicated when things won't sit still.

Pitfalls encountered since last post:

Altitude. I went up to almost 4000 feet 2 weeks ago and was greeted with a surprise. My power dropped to absolutely nothing when trying to climb hills. I happened to have my laptop so I plugged it in and ran a timing and fuel trace at the same time. Well lookee there...the vacuum isn't going over 80kPa. Huh...thats interesting...that ambient pressure is also 80. Lets think about this. The cars tuned to 100kPa and I only have 80 available to me. Well...I'm out of tuning. The computer has no reference point to go any higher. This isn't a problem in a normal car or even a less aggressive one as they idle at 2-11kPa...well mine is at 70kPa with spikes to 80 at idle. Right there I have less area to work with than a normal vehicle. When the other end of my chart got cut down to...I had no where to go. So the end problem here is a mismatch tune layout exacerbated by a decreased atmospheric pressure. This is all really "mechanical" (as opposed to chemical) problems. A carb overcomes this by just being sloppy (in relative terms). EFI overcomes this by having a large table to compensate. Another way it is dealt with is strictly chemistry. The O2 sensor sees that its running rich at elevation and starts pulling fuel out of the equation. The O2 sensor and computer doesn't know (or care) if its in Death VAlley or Mt Everest...its just knows its chemistry is wrong and adjusts for it. Since I no longer have a pressure (atmospherically) based tuning system I don't have a table that relies on pressure changes in motor to decide its fueling points. It puts the same fuel in at any elevation UNLESS the O2 sensor says its wrong then it will start modifying things for me behind the scene.

How did I get around this problem? I saved my current ECU file and recalibrated my table from 60 to 100 to 60 to 80. Its cheating, but it got me through the weekend. Now that I'm running TPS I have the full table operating and elevation changes are handled through chemistry not through shifts on my fueling table. Really....this is how it should be anyway.

The other pitfall had nothing to do with EFI theory or CB's kit. The A1 sidewinder I have has a leak in it. Its been in on/off use for a year and is still in very good condition. When running rich I noticed soot buildup on the muffler tip...which is fixed by dialing the fuel table in. Why is this important? The soot also acted as a marker in other way. It pointed out where the exhaust leaks are. I used high quality vanagon seals at all the flanges and put a trimill seal on the header to muffler joint. I also used exhaust putty to seal up all the joints in the header system. A1 said these measures were unnecessary. Perhaps in a carb setup where theres no feedback or calculation involved it may not matter, but with an O2 sensor downstream of the leaks...you can bet its going to kick in some lean readings. And it does. I've been battling spikes in my idle since day one. It causes my motor to run rich to compensate and then settle back down. Well, I disabled my IAC system to make sure it wasn't compensating for a super rich condition caused by a faulty injector (no bad injectors found, just a troubleshooting technique) and watched my AF meter on the CB dashboard. The car ran within points of my target AF then suddenly I would get a 16+ spike. As the motor settled you got a whiff of rich exhaust and a drop in the idle while it got back into its groove. So, I've located two leaks...one in the middle of the header collector and the other just at the main flange. These are both factory leaks. I'll be visiting A1 about this as this is a crucial part of the EFI system.

Mon May 07, 2012 12:44 pm

Thanks for the continued updates. I am keeping a close watch on this thread. The information about the system response at altitude is particularly interesting since we have a little cabin in the local mountains. Hopefully I will be tackling the motor and EFI by late summer or early fall. I ordered a SyncLink throttle control system from Jake Raby this morning.

Mon May 07, 2012 12:54 pm

Those pulley links are cool. I wish there was a way to use them with heaters still.

The altitude thing became an issue because I was using the wrong type of tuning base for the type of motor I have. If I went TPS from the beginning I wouldn't have had a problem. The computer would have known what to do with itself and I wouldn't have noticed much of a change. The solution is either a lesser cam, different exhaust, or use TPS from the start.

Mon May 07, 2012 1:05 pm

Yeah, all good stuff to know. The Spyder doesn't have heat so of course the fresh air ducts are not needed. I cut the fresh air ducts off my OEM doghouse shroud and dropped it off at the welder to have the unnecessary holes closed. I stumbled on to the SyncLink thread about the time I started following this thread.

I saw that Jake turned the carbs (or throttle bodies) around on his wife's vert w/AC and mounted the cables behind the fan shroud. He seemed to indicate that adjusting was a bear but doable. Of course he is using a Type 4 and a DTM cooling system. I found myself wondering if that would work on a doghouse shroud with heat. Not something I'll ever do, but something I was pondering.

We have a cabin in the local mountains. With a kid and two big dogs, I won't be taking the Spyder up there very often. On the other hand, it will be fun to drive on the mountain roads in the summer.

Mon May 07, 2012 1:19 pm

Just updated with problem number 2.

I don't think I will have any problems with altitude from here on now that I've switched over to what is called "Alpha-N" or TPS based fueling.

So far, nothing that I have run across in the install or tuning has been a regret or deal breaker. Even when I was feeling a bit hopeless chasing MAP tuning...CB walked me through it all and when it came time to throw in the towel on MAP they set me up with a good TPS table and I was ready to go just like that.

Tue May 08, 2012 6:28 am

I **ONLY** flipped the linkage to the back side of the shroud on Beth's vert to allow for her A/C compressor fitment. I would never, ever recommend anyone doing this for any reason, because it is a whore to manipulate. Thank goodness that after Synh-Link is set up it is virtually flawless and maintenance/ adjustment free.

Synch-Link doesn't matter which side of the shroud the linkage is fitted, you can even mount one wheel on the front side and one wheel on the back side if you want, we recently did that and it worked great. It was only done because the carbs had been modified and were missing an attachment point on one side.

On the Spyder, it doesn't matter because everything is easy to reach on both sides of the shroud.

Wed May 09, 2012 9:15 am

Pat told me a cool trick.

Put a cheapo plastic fuel filter in the vacuum line between the motor and the computer/fuel regulator "T". This does a major thing in 2 ways. It gives a CONSIDERABLY smoother vacuum signal. The first is, it acts as an accumulator like some braking systems have. It holds the vacuum stable longer so the computer gets a smoother signal. The second thing it does is slows down the change by forcing it through a paper element. Kind of like a gun supressor...it slows the air down so sudden spikes and pressure changes have a less sudden effect. Combine these two things together and you get a very good signal. I'm now only seeing 4 points different in vacuum fluctuations after this change.

DO IT!

Wed May 09, 2012 1:27 pm

Good tip. Unfortunately, I just got back from the parts store getting other odds and ends. I'll leave a spot for it and pick one up next time.

I've been following your build and appreciate all your posting. The more insight the better. I will turn the key tomorrow on mine for the first time.

Wed May 09, 2012 1:54 pm

I'd be ready for anything...from smoke and sparks to an easy startup. I'd have someone near the battery box to pull the plug incase something goes drastically wrong. I say this not out of experience or poor design, but out of the fear of cooking an expensive wiring harness from a pinched wire or a blown ECU. Don't forget to cycle the key about 5 times letting the pump autoshutoff each time. This will purge the air and let you check for leaks before you get a 45psi flame thrower out the back.

Did you remember to install the diode on the alternator field?

http://www.thesamba.com/vw/forum/viewtopic.php?t=510293&highlight=

Wed May 09, 2012 3:26 pm

I diddnt install the diode. The instructions said "some engines have run-on" so I thought I would wait and see. Maybe I should go ahead and do it?

Wed May 09, 2012 4:25 pm

I have seen most of these kits have the run on issue.

Here is another tip, if you have the software running, type (tool) while selected in the main dashboard. This will bring up the tool bar and allow you to turn the fuel pump relay on and off manually.

Tue May 29, 2012 8:06 am

Update:

I haven't been doing much tweaking on the system since I switched to tps based. I changed the af ratio and that about it. It's been running pretty good with nothing negative to report. I know I can squeeze a little more out of things, but have moved in to other projects for now that are taking up space.

Some new things. Do not run e85 on a standard tune. I did and it didn't want to start or hold idle. I didn't even hook my pc up to check he numbers...I immediately diluted it with 91 then things were better. It took me about a week to dilute it down to acceptable levels. The reason I did this? The new gas station was givin away 10usd free...why not I said.

I took the car over the cajon pass this weekend. It tops out at 4190 feet. I had zero problems. I think I'm running a little lean in the mid-low TPS level at 2500RPM..I can get some pinging by giving it more throttle under load. I've had this before under other highway conditions so it wasn't altitude related, but rather timing or fueling. Never the less, it got 21.5 MPG for the trip. Which is a little better than what the 1967 manual says I should get. This is a 2110 remember.

Mon Jun 25, 2012 9:20 am

Checking in with a tuning update...Its a little long...but believe me...its worth it.

Its been about a month now since the last update and I've got some fun and interesting news.

Early this month I attended the Classic VW show in Irvine and our good friends at CB Performance were there. I was swapping some stuff and asked if Pat could drop by and give my system a once over after the show cools down a little. So Pat dropped by and I whipped out my laptop and began to tinker. After about 30-40 minutes he had addressed a cold start issue, turned on the Decel operation (cuts fuel to almost zero during high rpm low throttle...coasting basically), adjusted some AF ratios, set my Accel operation to zero (more on this later) and a few other things. Went home happy. The car performed as expected, ran a bit cleaner it seemed, and didn't smell like kerosene when coasting (no wasted fuel on deceleration). Over the weeks it seemed fuel economy was up also.

So, this weekend. I decided to go into my fuel chart and modify it to basically zero out the learning history. This means adjusting the cell blocks by the amount the learning system is compensating for. Taking out or adding fuel based on a percent of change the computer has determined it needs. It sounds complicated, but its really not.

NOTE TO CB GUYS: Future software upgrade...an add or subtract percent option in the math buttons on the fuel table. It would make this type of tuning faster and easier.

I went on a weekend run to Santa Barbara, CA this weekend just for a get away and to pick up some parts. I spent about 1/4 tank getting to Camarillo, CA from Placentia, CA...84 miles. I can't use this for reference as I didn't refuel at this point and don't have any hard numbers outside of mileage. I got to our destination Carpinteria, CA and parked for the night. Next morning I started up and it was a hard cold start...not a nightmare...but not its usual self. I gave it a little blip of the throttle and it tried to die. I gave it a good bunch more...small cloud of blue smoke. Wonderful. Blown rings. That couldn't be it...nah. The engine lives a charmed life. CHTs under 350, good fuel, no racing or abuse, reduction boxes for torque...nah couldn't be it. Never seen it before either...oil level doesn't drop either. Ok..panic over...best I can see is during the long haul oil vapor built up in my extensive breather setup and dumped a few drops down the throttle body throat in the morning. Ok...mind at ease. That still doesn't explain the throttle blip and dying issue. So I went the rest of the day in Santa Barbara on egg shells and checking for smoke...nothing...oil level is fine.

I get home, to Carpinteria, and give the whole thing some thought. It acts like a carb with an incorrectly adjusted accelerator pump. I plugged in my laptop and watched the O2 readings at throttle "blip". Sure enough the O2 reads lean off the chart. AHA! Then I remember CB was talking to me about the Accel operation and that most motors don't seem to need it. After comparing my old tuning file and the post-classic one...sure enough...the Accel operation had been zeroed out. My "accelerator pump" had been "removed". I added numbers here until it blipped nice (good throttle response) then added a little more until I got some popping through the exhaust when blipped (too much fuel) dropped it back down a bit and hit the save button. Whadya know....the throttle tip in/blip problem went away.

I stopped in Camarillo, CA to pick up some stuff and fuel up. I check the mileage of the route home from the gas station, 84 miles. The ride home was uneventful and drove as expected. I finally get back to Placentia, CA and stop into the local Chevron to top off and get the results of my mileage test. *DING* 3.043 gallons. Hmmmm...84 divided by 3.043...27.6....

****27.6MPG****

This is on a CB Performance Magna EFI powered 2110 Type 1 with Hydraulic 110 cam (theres some debate whether its closer to a 130, but thats another Oprah) and A1 sidewinder Exhaust w/ Super Trapp arrestor for noise. The motor is installed in a 1967 Type 2 Riviera with full camper kit, 3.88 RP .82 4th, Redux boxes, and 6.70 R15 Bias Ply tires. The camper was fully loaded with 2 adults and one 110lbs dog, gear, and extras from pickups, also a roof rack with some large draggy objects in it. There was a strong patchy wind from the west. There is a large climb into Malibu/LA. Wikipedia describes it as "The grade is about 2.7 miles (4.3 km) and posted as a 7% grade—which translates as about one thousand feet of elevation change in less than three miles (70 meters per kilometer)." I could almost do it in 4th gear, but to keep things happy I dropped to 3rd.

Stick that in your Suby and smoke it.....

..Seriously though...

The CB Magna EFI system allowed me to diagnose and rectify problems on my own with accuracy and reliability. There was no "seat of the pants dyno" or pulling plugs needed. Just a basic understanding of what was going on and what could possibly fix it. This mileage number was also not adjusted for. Meaning, I didn't use my learning adjusted fuel table. I felt resetting the system back to before that point and then making changes would narrow possibilities and make solving the problem easier. With this in mind and the other factors working against efficiency in this case (load, hill climb, etc)...I feel 30mpg is not an impossible goal. Pat at CB also recommended a possible considerable MPG gain by changing injectors to a different spray pattern. Between tuning, driving, load, and parts...it may even be possible to push it all to 32 maybe 35.

I couldn't be happier with the system, the service, or my car.

Mon Jun 25, 2012 12:21 pm

Sniperx wrote:
NOTE TO CB GUYS: Future software upgrade...an add or subtract percent option in the math buttons on the fuel table. It would make this type of tuning faster and easier.

We use the multiply function for this. multiply by .99 is subtracting 1% multiply by 1.01 is adding 1%

I'm glad you are happy with the system

Mark Lawless
CB Performance

Mon Jun 25, 2012 12:48 pm

OH YEAH!!!! I did it the hard way........