We will be posting documents and images from various projects we are working on. Our projects typically include upgrading components in German automobiles for the purpose of bringing new and advanced functionality to older vehicles. Some examples include our upgrade kit for the Mercedes climate control system, known as the ACCII and our UTCIS line of replacement and upgrade Warm Up Regulators for Bosch CIS K-Jetronic systems. These systems are found on Mercedes, Porsche, BMW, Audi, Volvo, Saab, VW, Ferrari and others.
We’ve had reports that there is a difficult fit in some years of Porsche 930 applications. The fit problem arises from an interference between the right side of the UTCIS body and one of the intake runners. The UTCIS body is about 2.5mm too wide at this point. From what we can tell, this only occurs when the UTCIS is mounted flat on the intake manifold, without the spacers.
As of yesterday, we’re now shipping two sets of mounting bolts for the UTCIS. We’ve always included a set of M6 x 50mm bolts which bolt the UTCIS body directly to the stock mounting bracket. Mounting the UTCIS in this way puts the fuel fittings about 25mm lower than the height of the stock WUR. Previously we’ve noted in the manual that you can use 80mm mounting bolts instead to keep the fuel fittings at the stock height. We always included the spacers to do this, but not the longer bolts. Now we include both sets. Raising the UTCIS body also gives you a little wiggle room to fit the UTCIS between the intake runners on the 930. Please let us know if this is still a fit problem.
We planning to reduce the width of the UTCIS from 65mm to 60mm anyway, so the fit overall should be easier. This change is scehduled to take effect as of November 15.
V12 applications are now supported! Both Lamborghini and Ferrari produced V12 cars which used 2 each of independent 6-cyl CIS systems. We just completed a successful installation on a Lamborghini V12 using 2 each of UTCIS-V systems. The shop who purchased the systems over a year ago for a different car sent us his units to be upgraded to the new G7 version. We upgraded the systems, cleaned them up, tested them, and shipped them back. The car is now happily spinning its tires with a happy owner prowling the streets of SoCal. If you have a V12 application you’ve already noticed that the Warmup Regulators are NLA. Now there’s a programmable digital replacement that’s a simple drop-in.
We have at long last released the new version G7 for shipment. For those of you who have been waiting patiently — thank you. The first step in this release was to finalize the machining of the new body which replaces the OEM WUR. The other half is a new controller which handles the control pressure in a faster and more precise way:
Version G6 introduced a mechanical pressure sensor and the sensor output is part of a feedback loop. Putting the sensor in the feedback loop had a number of disadvantages. First of all, the sensor is a mechanical device and it’s slow. It takes about a 1/10 sec to get an accurate reading and several readings have to be averaged in order to get an accurate result. Also, when the Control Pressure is lowered, for example when the accelerator pedal is mashed, then the control pressure has to be measured and the results fed back into an adjustment of the injector on-time. All of this takes time, which caused the response time to be less than optimum.
We first found this effect when an owner/driver with a 930 race car reported that when he came out of a hard turn and got on the throttle, the feedback loop would not catch up until he was ready to snap off the throttle again. His was a race application, but responsiveness without delay is a great feature for any modern car. Our belief is that these CIS cars can run as well or better than an EFI car, without the pain or cost of an EFI conversion. For that to be a reality the loop response time had to be reduced or eliminated.
The G7 version takes the CP sensor out of the feedback loop. When the engine control map sets the control pressure the response is now instantaneous. The engine control map sets the control pressure using Manifold Air Pressure (UTCIS-V) or a combination of MAP and RPM (UTCIS-PT). When the car is started the G7 version will wait until the car is idling and then will apply a calibration value to the duty cycle of the injector. In this way the CP pressure sensor is used as a long-term trim, not an active part of a feedback loop. This also allows us to reject the sensor value if it breaks (out of range or far from previous value). That’s an important feature because an injector is a far more reliable component than a mechanical pressure sensor!
So far we’ve found that a calibrated injector is slightly more accurate then the output of the pressure sensor. If this observation holds up over time then at some point the pressure sensor may removed from the product.
If you have firmware 3.73 or higher your UTCIS can be upgraded to the current G7 firmware at no charge. You may have heard that there have been delays with upgrades in the past. This has been true in most cases because there has been a bottleneck in the production of new valve bodies. Starting today there is no delay in the upgrade because the older model can be upgraded with a firmware change only without adding a pressure sensor using a calibration of the fuel injector. If you have an older UTCIS with a stepper motor instead of an injector then you will require a new valve body with an injector. The cost of the the valve body upgrade is $150.
It’s been awhile since the last update regarding the UTCIS, but it hasn’t been for lack of progress. The progress has been slow but steady. The goal of the UTCIS is to quickly and easily upgrade the CIS injection system to the same level of precision, reliability, and tune-ability as a moden EFI car, but without the pain and cost. We think we’ve finally got it. Many thanks to those of you who have helped us and encouraged us to keep up our support of these older cars. We now have results we can show you. Both the electronics and the mechanical design have changed. Here’s what the UTCIS looks like now:
The controller is quite a bit smaller and simpler, but faster and more capable.
You can expect regular updates from us this week and next week. For those of you who have been patient with us, or more than patient, we thank you. This version has been performing extremely well and the next few posts will be focusing on features of the new version and how to use them.
With the new UTCIS in production and working well on our 924 Turbo it was time do some heavy duty tuning with the UTCIS-PT. With exceptionally good timing, Chris at Turbo Kraft in Phoenix generously offered to let us be present for some dyno tuning on a beautiful 1981 Porsche 930 with UTCIS-PT installed. The car is already making close to 400hp at the wheels but suffers from a very rich mid-range and a lean top end.
With a new version G3 UTCIS-PT in hand the guys at Turbo Kraft had it hooked up and working in no time.
After some PC issues, solved with a ground wire and a reboot, the tuning was underway. The very skilled tuners at Turbo Kraft started tuning at the low end setting the idle to about 14:1 AFR by raising the control pressure everywhere 2000 rpm and under. With the idle ironed out Chris moved on to the mid-range and then up to full power/boost runs.
It was immediately obvious the edits made in the UTCIT-PT engine maps were having a direct effect on AFR. The mid-range started out very rich, around 10:1 on a few runs. Chris raised the control pressure and brought the AFR up to around 12:1 in most of the mid-range. The full power/boost runs showed that above 5000 rpm the engine was starting to go lean as RPM increased. The control pressure was then bottomed out to 0.8 bar above 5000 rpm at full boost. This helped lower the AFR but it could still be a little richer so there may be a max fuel delivery problem. At this time our dyno time expired for the day but Chris said he wanted to spend some more time tuning the UTCIS-PT. When Chris is finsihed I will post some before and after dyno plots so check back soon. Overall it was a very successful tuning session and we would like to thank Chris and the rest of the guys at Turbo Kraft for letting us observe.
Initial testing on our Porsche 924 Turbo with the new generation UTCIS (what we are calling G3) is now complete. We have started production and are FINALLY ready to ship the product! We have made extensive improvements with this version – many are described here in this post.
Improvements made in the new UTCIS
As anticipated, the new version is much faster on the car. Control pressure responds to manifold air pressure and RPM (PT only) almost instantly. Below is a video showing the control pressure responding to manifold air pressure as the throttle opens and closes on the 924.
Control Pressure Range
In the previous version of the product we had to limit the regulation range of control pressure from 2.0 bar to 4.0 bar gauge. In the new version, we have increased the regulation range. The control pressure is now settable anywhere between 0.8 and 5.0 bar. In many applications this expanded range is not necessary, but to satisfy the very fuel thirsty high boost cars it becomes very important. There are also a few CIS equipped models out there that operate with normal control pressures outside the previously used 2.0 to 4.0 bar range.
The expanded range was not possible in the previous version of the product. The system did not have enough resolution above 4.0 bar and at values below 2.0 bar, the stepper motor was too slow to keep up. In this new version we were able to accomplish a wider range and more speed with the combination of a better stepper motor and smarter software. The stepper motor is not only faster but also has over four times the resolution. This allows us to move the valve faster and increase precision at the same time. In the previous version of the product the motor moved at the same speed regardless of control pressure needs. The new software now measures the difference between the target and the current control pressure and then dramatically increases motor speed if the difference is greater than 0.5 bar. This allows the product to fine tune control pressure under static conditions like idle and cruising but also scream to high and low control pressure values when needed, for example conditions such as hard acceleration and hard deceleration.
The new motor also uses ball bearings at both ends rather than bushings and the shaft diameter is up from 3mm to 5mm. The available torque from this motor is far greater than the product requires and allows us to run in at a fraction of its rated output. We also implemented a current limiting driver which helps keep the motor cool and running strong even if the vehicle battery voltage fluctuates up and down.
The motor comparison is shown below:
The new version of the product is also more rugged. The motor spacer and standoffs were eliminated and the motor now sits and seals directly on the valve body. All moving parts are fully enclosed, lubricated for life and sealed with in the aluminum valve body.
The valve body is now easily separated from the electronic controller case if needed thanks to a transition wiring connector board and switching to high pressure braided tubing secured with fuel injection style hose clamps.
Sensors & Control Logic
The method for determining engine temperature and warm-up enrichment requirements has also changed. In the previous version the temperature was measured inside the controller case and a heater was turned on once the engine was running. Engine temperature was measured indirectly by sampling the temperature inside the engine compartment plus the added heat from onboard heater. In the new version we now measure engine temperature directly by measuring the temperature of the motor attached to the valve body which is bolted to warm place on the engine. To keep wiring to a minimum we came up with an innovative way of measuring the motor temperature so there is no need for an additional temperature sensor. An added benefit is that the controller does not have to be mounted in the engine compartment although it still can be. Measuring engine temperature this way will also provide even more consistent warm-up regulation and quality. The warm-up table is of course still fully programmable.
The pressure sensors used in the previous version of the product are no longer in production which is a BIG reason for the long delay in releasing the new version. In order to cut down on part count and development time for a replacement sensor we made the cost leap to Freescale pressure sensors that are fully calibrated and compensated on board the sensor. This adds an additional layer of reliability. In the rare of event of a controller crash the old sensors had to be re-calibrated and the only way to do that was here at our facility. In this new version the sensors are pre-calibrated by Freescale and do not require any input from us. A very stable external memory chip has also been added so in the event of a controller crash a simple reset performed with in the UT Engine Management software will restore everything to factory defaults!
Move over serial, here comes USB
The other major change is conversion to USB! For all of you out there with modern laptops and no serial port you can now connect directly via USB. The other MAJOR benefit to this is that USB carries its own power from the PC to the controller. This means the UTCIS does not require any external power or connections besides USB to establish communication. All re-flashing, programming, and editing of the UTCIS can now be done on the bench or on the car once the included USB cable is connected to a PC. This also means the engine does not have to be running during programming and re-flashing although it can be.
In the PT version, the RPM section and measurement method has been drastically improved. There is no more calibration required and the number of cylinders is now settable in the UT Engine Management software so the UTCIS will function correctly on 4, 6, and 8 cylinder engines. The RPM signal is now measured only at the coil so there is no danger in killing the tach signal. The new method is also compatible with aftermarket ignitions including MSD and Crane.
New to this version is an added feature that allows the output of an oxygen sensor to be connected to the electronic controller. It will read in a raw narrow band oxygen sensor or it can also read a wide-band oxygen sensor signal scaled between 0 and 3 volts. Most aftermarket wide-band controllers such as this one:
available from Innovate Motorsports have extra scalable outputs that can be used by the UTCIS controller. The UTCIS does not use the oxygen sensor to regulate control pressure. The value is simply displayed in the UT Engine Management software for the purposes of tuning. UT may add additional features using this input in the future, though no explicit plans exist at present.
Data Logging on its way
We are also currently expanding the data logging feature in the PC software. This will allow for dyno runs or pulls through the gears and then go back later and look at where the control pressure needs adjusting. The data logging feature will record control pressure, manifold air pressure, RPM (PT Only), O2 sensor output for a finite period of time. This feature is still in the works but when it becomes available it be a no charge upgade to the UT Engine Management PC software. No changes will be required to the UTCIS itself to enable data logging.
UnwiredTools would like to thank everyone for thier support and patience in the new release of the UTCIS. Circumstances outside our control have pushed the release well beyond our initial predictions but we are extremely happy with the final result and we know customers will be as well.
After many delays and false starts the next generation of mechanical fuel injection is ready. The UTCIS is finally getting back into production. Your patience and support has been sincerely appreciated. The past few months have been interesting to say the least.
The UTCIS uses two pressure sensors, one is used to measure the control pressure, the other MAP. Both sensors were uncompensated, unamplified types which we surround with proprietary electronics to make them work. This approach was more complicated than an off-the-shelf sensor but it gave us advantages in size, power, and cost. In September the manufacturer of he sensors ran into financial trouble, reorganized, and discontinued our sensor. The company was still in business so we thought no problem, we’ll change the design the use a different model. No sooner was the re-design complete that this sensor was no longer available as well. OK, we got the message. An off-the-shelf fully integrated sensor from Freescale was used instead. As an aside, Freescale has been doing relatively well because a lot of small sensor vendors are consolidating and the remaining vendors are picking up the business.
The aluminum body of the UTCIS has a lot of machining which we sub-contract to a very capable machine ship in Phoenix. Unfortunately they went bankrupt in October. Our search for a replacement vendor brought us to a larger and better equipped shop in Prescott. Unfortunately they were swamped with business from smaller shops who had left their customers scrambling. Our new machine shop is great but then we had a delay because they are so busy.
We had been using a motor vendor in China for the stepper motor. Given all of the other changes and delay we took the opportunity to upgrade the stepper motor to an aircraft grade motor.
Shown below is an interim solution with a Nema 11 Stepper motor only available from a supplier in China. Just last week we decided against using this motor below.
Shown below on the left is the original motor we started with and on the right is newest Nema 14 motor available from a US supplier. We just switched to the Nema 14 stepper motor which offers more torque and better availability than the Nema 11 (above) from China.
Finally shown below is a computerized rendering of the new body with the recently selected high quality Nema 14 motor.
Below is a Braided Reinforced fuel line which replaces the Teflon line used on the old version.
The UTCIS body is now easily separated from the controller if needed.
We will be shipping a small quantity of units the first week of next month with the balance of our production coming by the end of April. Thank you for your business, encouragement and continued support.
In the last post http://unwiredtools.wordpress.com/2008/11/26/utcis-pt-install-porsche-924-turbo-931/ we left off with a much more responsive, better running 924 Turbo after some minor tweaking on the UTCIS-PT base map. The map saved into the unit held the control pressure at a constant 3.0 bar for all RPM’s and manifold air pressures. Shown below is a screen shot of the UT software and the constant 3.0 bar MAP.
The above control pressure MAP in the 924 Turbo is very safe and works quite well. It makes good power but the AFR is probably richer than it needs to be. After running this map for a full tank of gas a fuel economy of 26 mpg was obtained with about 80% freeway driving and a fairly heavy foot. While we are waiting for the intercooler to arrive we decided to find out what type of maximum fuel economy is possible just by editing the map in the UTCIS-PT. The first MAP designed for good fuel economy that we tried is below:
In this control pressure MAP the control pressure gets slightly higher once the engine is off idle and under load which leans out the mixture. Once boost starts to come in the control pressure goes back down to richen up the mixture. This keeps the engine safe under boost but lets us run the engine slightly lean in the mid-range where the engine spends most of its time while cruising. Again this control pressure MAP is used at all RPM’s. After a full tank of gas with the above control pressure MAP and very similar driving conditions a fuel economy of 32 mpg was obtained. There was also no noticeable difference in driveablitiy or performance. So far the results are impressive but we are not quite done tweaking the map. The engine should handle running a little bit leaner in the mid-range which should result in even better mileage.
It has taken a few months to get our 1980 924 Turbo on the road and driving comfortably but it is almost there. There are still cosmetic issues to take care of but other than that the car is running and driving like new. We learned a lot in the process and would like to share some of that info and also discuss the install and first round tuning of our UTCIS-PT.
There was a rather long list of items in need of attention on this car when we started. The car has close to 100,000 miles on it and a lot of maintenance issues were ignored by previous owners. The two most obvious issues were very worn (non-existent) rear shift bushings and a mangled clutch master cylinder mount. Fixing the clutch master cylinder mount is covered in another post:
According to the Haynes manual, the transmission must be removed in order to replace the rear shift bushings but in order to save some time we decided to find another way. We ended up cutting two access panels under the rear seat and replaced the bushings without removing a single fastener on the transmission! This procedure will be covered (with pictures) in a future post from UnwiredTools so check back soon.
Now on to the engine and fuel management. Our goal initially for this car is to return the engine as close as possible to stock form. The 1980 924 Turbo US model came from the factory with a de-tuned (compared to the European model) 1984cc 4 cyl engine fed by a K-lambda CIS fuel management system. This engine makes 150hp at 5500 rpm according to the Haynes manual. The European model makes an additional 20 hp by running a higher maximum boost pressure. Future plans for our US model include an intercooler, increasing boost, programmable ignition and of course a UTCIS-PT. According to the math we should be able to achieve 200+ hp with these upgrades and everything will be covered in detail in future posts. The engine was running and starting okay before we got our hands on it but it became obvious among other issues that the lambda loop was not functioning. To avoid any confusion the oxygen sensor was replaced immediately and upon removing the old sensor we realized it was not connected to anything. The green shielded wire that normally connects to the oxygen sensor was cut and just hanging inside the engine compartment.
Before reconnecting the green wire to the new oxygen sensor we ran a few tests to ensure the lambda control unit was still alive and functioning. Using a few alligator clips and 1.5 volt AAA battery it is very easy to stimulate the lambda control unit and see if it responds correctly. There is a three pin test connector inside the engine compartment that when probed with a duty cycle meter or scope shows the signal being sent to the lambda frequency valve. On our car the black wire on the three pin test connector is the signal wire. Once the engine was fully warm we grounded the green oxygen sensor wire coming from control unit to simulate a lean condition. The frequency valve duty cycle responded correctly and went to 95% in order to richen the mixture. In the next test the 1.5 volt battery was connected to simulate an overly rich condition. Again, the frequency valve duty cycle responded correctly and went to 10% in order to lean out the mixture. From these tests we were able to conclude the lambda control unit was functioning correctly. Something interesting we noticed is the amount time it takes for the duty cycle to swing from end to end. It takes about 15 seconds for the lambda control loop to swing the duty cycle from full rich to full lean or vise versa. It is actually the job of control pressure regulator to make rapid changes to the air/fuel ratio in response to engine load. We could now connect the new oxygen sensor to the lambda control unit via the green shielded wire. For a very comprehensive explanation of how to perform these tests pick up How to Understand, Service, and Modify Bosch Fuel Injection and Management by Charles O Probst or feel free to give UnwiredTools customer service a call.
With the new oxygen sensor connected it was time to check the CO screw adjustment. There are few ways to do this. We chose to measure the duty cycle of the lambda frequency valve at the test connector and adjust the CO screw until the duty cycle hovered right around 50%. With the CO screw adjusted the engine started running much smoother at idle.
The engine now appeared to be idling very well with a fully functioning lambda control loop. As a double check an air/fuel ratio gauge (24$ from AutoZone – a very good investment) was installed in the dash in order to keep an eye on the oxygen sensor output.
As expected the gauge showed the lambda loop was working correctly by bouncing from lean to rich. Finally, we were ready to install the UTCIS-PT unit in place of the stock control pressure regulator. Before removal of the stock unit we decided to take a quick look at control pressure using the UnwiredTools CIS control pressure gauge. It takes less than minute to install the gauge so it was worth a quick look.
Not too surprisingly, like other cars of this era, the stock control pressure regulator had failed. The warm engine control pressure measured very close to system pressure, around 80 psi (5.5 bar). At idle on this car with the vacuum line connected the control pressure should be around 43 psi (3.0 bar). Adjusting the CO screw earlier compensated for the failed control pressure regulator at idle but in cold start conditions and high engine output, lean conditions would have arisen. Lean conditions, especially in a turbocharged engine under boost, will quickly destroy the engine. Not yet a dyno tested theory but very high control pressure would also cause an increased pressure drop across the air metering plate. This would lead to some loss of power at the top end in addition to the lean mixture.
Removing the stock control pressure regulator was not too difficult with the coil out of the way and a 5mm hex drive attached to a ratcheting wrench.
The lower mounting bolt is much easier to remove with this tool setup. Removing the fuel fittings first before loosening the mounting bolts is a good idea. Installing the UTCIS-PT was much easier than removal. The fuel lines mount to the UTCIS-PT body in same locations as the stock unit. The same vacuum line removed from stock control pressure regulator connects right back up to the new UTCIS-PT. The electrical connector was a bit of pain to remove from the stock unit but plugged right in to the UTCIS-PT.
After mounting the controller on the firewall and securing the wire harness the install was complete. The total removal and installation time took about 45 minutes. The engine fired right up and the UTCIS-PT was communicating correctly with the laptop. As an initial baseline the control pressure map was set to 3.0 bar (43.5 psi) at all rpms and manifold pressures using the UT engine management software. As expected the warm engine was now running rich because the CO screw was adjusted earlier to compensate for very high control pressure. A quick adjustment of the CO screw brought the frequency valve duty cycle back to hovering right around 50%. The afr gauge was now cycling back and forth from rich to lean indicating a properly functioning lambda loop and an average air fuel ratio right at stoichiometric.
It was finally time for a test drive and what an improvement! The throttle response, low end and top end power were all noticeably improved. The idle is butter smooth compared to before and the popping coming from the exhaust on deceleration is much less noticeable. As the engine load and boost increase the afr gets progressively richer. At full boost and wide open throttle the mixture stays planted rich and provides an exciting push from the back of the seat. Gas mileage should also be improved and acutal mpg numbers will be posted in future posts. The intercooler is on the way and pictures/updates of the completed install will follow shortly.
I had a very nice conversation with Dave Newfield of “In Your Face Gauges” last week. Dave described how he takes a stock 930 tach and adds some very cool features. I could describe it in words, but video is much better. Have a look at Dave’s You Tube posting:
Many UnwiredTools UTCIS enthusiasts have expressed interest in using Dave’s gauge setup with the UTCIS-PT. We’re looking into it.