Have to say, I've not seen this before and can't think of a reason why someone would want to do it, but it certainly is unique!
Sunday, May 20, 2007
Maserati 3200 GT with Gull-Wing Doors?
I was just checking the second hand Maseratis for sale and came across this...
Have to say, I've not seen this before and can't think of a reason why someone would want to do it, but it certainly is unique!
Have to say, I've not seen this before and can't think of a reason why someone would want to do it, but it certainly is unique!
Saturday, May 19, 2007
Intercoolers
I've been doing a bit of research on intercoolers having recently found out a couple of things.
First was from Buddy at Wereld Auto who informed me that intercoolers from a VW Golf mark IV TDi will fit on the Ghibli, although the one manifold has to be removed and a new one constructed and welded on its place. The good thing about this solution is that it's cheap (these intercoolers can be purchased on ebay for less than €100) and they offer good performance since they're twice the thickness of standard Ghibli units. However, I don't like the idea of having to modify them.
The second option was hinted to me by Jorrit at Auto Forza. He informed me that the intercoolers on the V8 engined Quatroporte is about 30% thicker than the units on the V6 cars. I then happened to stumble upon a picture of a 3200 GT being sold for parts and noticed the intercooler... the location is identical to that of the Ghibli and Quatroporte and the configuration of the ports looked the same.
So I wrote to EuroSpares and enquired as to whether they had any second hand 3200 GT intercoolers and if they could check whether they fitted a Ghibli. They wrote back to inform me that they'd been informed from the factory that the Ghibli intercoolers will fit the 3200 GT (wrong way 'round but by definition, the 3200 GT intercoolers must fit the Ghibli (although the ducting may require a bit of modification)).
Above is the parts schematic from the 3200 GT. Compare the size of the intercoolers with those for the Ghibli in the schematic below.
So I'm tempted to try and source some 3200 GT intercoolers and get Jorrit to try fitting them when he's fitting my new bumber (when it arrives... it's been on order from the factory for about 2 months now!).
As usual... I'll keep you posted of any updates.
First was from Buddy at Wereld Auto who informed me that intercoolers from a VW Golf mark IV TDi will fit on the Ghibli, although the one manifold has to be removed and a new one constructed and welded on its place. The good thing about this solution is that it's cheap (these intercoolers can be purchased on ebay for less than €100) and they offer good performance since they're twice the thickness of standard Ghibli units. However, I don't like the idea of having to modify them.
The second option was hinted to me by Jorrit at Auto Forza. He informed me that the intercoolers on the V8 engined Quatroporte is about 30% thicker than the units on the V6 cars. I then happened to stumble upon a picture of a 3200 GT being sold for parts and noticed the intercooler... the location is identical to that of the Ghibli and Quatroporte and the configuration of the ports looked the same.
So I wrote to EuroSpares and enquired as to whether they had any second hand 3200 GT intercoolers and if they could check whether they fitted a Ghibli. They wrote back to inform me that they'd been informed from the factory that the Ghibli intercoolers will fit the 3200 GT (wrong way 'round but by definition, the 3200 GT intercoolers must fit the Ghibli (although the ducting may require a bit of modification)).
Above is the parts schematic from the 3200 GT. Compare the size of the intercoolers with those for the Ghibli in the schematic below.
So I'm tempted to try and source some 3200 GT intercoolers and get Jorrit to try fitting them when he's fitting my new bumber (when it arrives... it's been on order from the factory for about 2 months now!).
As usual... I'll keep you posted of any updates.
Friday, May 18, 2007
Maserati Ghibli V8
Today I had a holiday and so took the opportunity to visit Buddy at Wereld Auto.
I went there to find out more about the KMS, but spent most of the time discussing Ghiblis. He had two in the workshop. one was a MY93 series 1a (but had Mistral style wheels). It was having some problems with a seized turbo, but Buddy has been having trouble trying to source a replacement. So he's working on manifolding a small Garret turbo; designed to flow 180 hp and with a good bearing package, it should be spooled up by 2000 rpm and offer very good low rpm drive-ability (which is what the owner of the car is after).
Buddy says that if he can get the Garret turbo mounted satisfactorily onto the Ghibli's exhaust manifold then he plans to make a kit that will offer a bolt on replacement for Ghibli owners. He also says that by using the Garret turbo, there's lots of scope to play around with compressor sizes to open up tuning possibilities. He thinks he can come up with a kit for around €800 per unit, which is far cheaper than getting a Maserati original part! I'll keep you posted if he managed to pull it off!
OK... the other car was a a very special MY92 Ghibli, but with a Maserati Evoluzione engine and running gear from a Quatroporte, running on a KMS ECU and with larger intercoolers (actually from a Golf Mark IV TDi... they're twice as thick as standard Ghibli units, but the bottom (outlet) manifold needs removing and a new one welding on). Buddy took me for a drive in the car... I've not had that much of a buzz since being in the passenger seat with Top Gear's " The Stig" at the wheel. What a fantastic performing car! The brakes and handling could be improved, but the engine performance is very impressive indeed... the car felt like it had a huge normally aspirated engine, with bags of torque available at low revs... no having to wait for the turbos to spool up like on my Ghibli or having to drop the gears to get the engine revving in it's power band, just stick your foot down and away you go! Buddy's not had it on the dyno recentyl but thinks the performance is around 380 hp with a little under 600 N.m torque. He's set up the KMS pretty conservative so as not to stress the engine too far and make the car very driveable... for example the boost pressure is mapped against throttle position to avoid the big spikes in torque when accellerating out of corners that the car was previously delivering and catching the owner off-guard. It seems that Buddy has a fair bit of track driving experience and he said that the torque spikes were occuring too quick for him to reliably control.
Very, very impressive... kind of gets me tempted to try and source a Maserati V8 and have it fitted in my car! It's a bit tight in the engine bay, but it fits ok and there's actually more ground clearance underneath the car since the V8 has a shallower sump. Plus, since this Ghibli was registered in '92 it doesn't require cats which helps out too.
Regarding the KMS... Buddy still needs to find out if it can run an odd fire engine, so i promised to send on the info that I've acquired and he'll ask the manufacturer. So, lots to report back, but not much yet on the KMS!
I went there to find out more about the KMS, but spent most of the time discussing Ghiblis. He had two in the workshop. one was a MY93 series 1a (but had Mistral style wheels). It was having some problems with a seized turbo, but Buddy has been having trouble trying to source a replacement. So he's working on manifolding a small Garret turbo; designed to flow 180 hp and with a good bearing package, it should be spooled up by 2000 rpm and offer very good low rpm drive-ability (which is what the owner of the car is after).
Buddy says that if he can get the Garret turbo mounted satisfactorily onto the Ghibli's exhaust manifold then he plans to make a kit that will offer a bolt on replacement for Ghibli owners. He also says that by using the Garret turbo, there's lots of scope to play around with compressor sizes to open up tuning possibilities. He thinks he can come up with a kit for around €800 per unit, which is far cheaper than getting a Maserati original part! I'll keep you posted if he managed to pull it off!
OK... the other car was a a very special MY92 Ghibli, but with a Maserati Evoluzione engine and running gear from a Quatroporte, running on a KMS ECU and with larger intercoolers (actually from a Golf Mark IV TDi... they're twice as thick as standard Ghibli units, but the bottom (outlet) manifold needs removing and a new one welding on). Buddy took me for a drive in the car... I've not had that much of a buzz since being in the passenger seat with Top Gear's " The Stig" at the wheel. What a fantastic performing car! The brakes and handling could be improved, but the engine performance is very impressive indeed... the car felt like it had a huge normally aspirated engine, with bags of torque available at low revs... no having to wait for the turbos to spool up like on my Ghibli or having to drop the gears to get the engine revving in it's power band, just stick your foot down and away you go! Buddy's not had it on the dyno recentyl but thinks the performance is around 380 hp with a little under 600 N.m torque. He's set up the KMS pretty conservative so as not to stress the engine too far and make the car very driveable... for example the boost pressure is mapped against throttle position to avoid the big spikes in torque when accellerating out of corners that the car was previously delivering and catching the owner off-guard. It seems that Buddy has a fair bit of track driving experience and he said that the torque spikes were occuring too quick for him to reliably control.
Very, very impressive... kind of gets me tempted to try and source a Maserati V8 and have it fitted in my car! It's a bit tight in the engine bay, but it fits ok and there's actually more ground clearance underneath the car since the V8 has a shallower sump. Plus, since this Ghibli was registered in '92 it doesn't require cats which helps out too.
Regarding the KMS... Buddy still needs to find out if it can run an odd fire engine, so i promised to send on the info that I've acquired and he'll ask the manufacturer. So, lots to report back, but not much yet on the KMS!
Friday, May 11, 2007
ECU Upgrade
Following a lead from Jarle in Norway, I started to investigate Kronenburg Management Systems as an option for upgrading the engine management system on my Ghibli to overcome the limitations of the OEM system installed by Maserati.
The Magnetti Marelli system installed by Maserati is very old compared with today's technology. To put it into context, this system was installed in Ford Sierra Cosworth cars back in the late eighties and early ninetees. The clock speed for this system is around 4MHz. Compare this to a modern laptop or desktop computer, which runs with a clock speed an order of magnitude quicker at 4GHz. In fact, you'd be luck to find a microprocessor that runs less than a hundred times quicker than that utilised in the OEM engine management system.
The effect of this is that the engine management system currently installed in my Ghibli makes significantly less decisions per second than the systems installed on typical modern street cars. Things happen and change very quickly in a car engine, particularly when it's reving at 7000rpm. That's over 100 revolutions of the crank every second, which is 25 detonations per second per cylinder.
The way an engine management system works is by taking a snapshot of the engine, which means recording the values of certain engine parameters at a moment in time by reading the output signals being sent from the various sensors located around the engine. The key sensors on the Ghibli are the inlet air temperature and the inlet air pressure for example. In order for the engine management to make decisions, it needs some reference data. This reference data comes in the form of look-up tables, which is just how people used to make complicated calculations in the days before electronic calculators. These are called maps.
A 2 dimensional map allows you to look up an answer if you know two numbers. Let's say that the inlet temperature is listed across the top row of the look-up table and the inlet pressure is listed along the left column of the table, the answer can be found by drawing a line down from the top inlet temperature value and a line across from the left inlet pressure value. Where the two lines cross is where the answer lies.
The answer tells the engine management system to perform a certain task. However, this answer is only applicable to the moment in time when the two input values were recorded, so if the system is too slow, by the time it carries out the response, it will no longer be applicable... a delayed response.
If there are more sensors, then the management system needs more look-up tables in order to know what decision to make, which takes more time. Also, the decision may require more than one response... for example the amount of fuel that is required and the time at which to ignite the fuel may need to be adjusted. Each additional response takes additional time.
My point here is that the quicker the management system can make decisions and respond, the more effective it is. In the Ghibli, it's obvious that the speed of the engine management system is not optimal, simply by virtue of the fact that TWO management systems are required, one to control the amount of fuel being injected and one to control when the fuel is ignited. So even when the Ghibli was being manufactured, it seems to have been recognised that processing speed was a limiting factor. Therefore, increasing processing speed should have a significant effect on the performance of the engine.
The next issue with the OEM management system is its lack of diagnostic capability. The only real way of determining if the system is making the right decisions is by measuring the performance of the engine, either by having a feel for the car or by sitting it on a dyno. The system does not have an easy way of monitoring what decisions are being made, except by taking the car to a specialist who can tap into the system and monitor what it is doing, which requires a degree of skill, technical knowledge and the appropriate equipment.
Modern management systems however offer a convenient way of monitoring what is happening. They generally have a convenient port into which you can plug a computer and with special software, you can record and review what decisions are being made.
Finally, if you recognise that sub-optimal decisions are being made, you need to be able to change the decision making process, which generally means altering the look-up tables or maps. Again, on the OEM management system, this is not a trivial thing to do and requires removal of the system and some special hardware to reprogramme the memory chips on which the data is stored. Also, the way this data is stored is not very user friendly, as it is stored in a way that is convenient for the microprocessor to read and not the car owner.
The beauty of a modern system is that all these things are possible. With modern software, you programme the data in a form that is easy to understand (using decimal numbers instead of hexidecimal or even binary code required for older systems) using a laptop computer. You can also record data from the engine, review it at a leisurely pace, determine what changes you want to make, update the look-up tables and see straight away the results... without having to go to a chip-tuner!
So, after contacting Van Kronenburg and discovering that they're a Dutch firm (much to my delight, since it means that they're local), I was a bit downheartened that I never heard back from them and it is then that I ordered the PLX Devices R-500 datalogger. However, I subsequently received an email from Buddy at Wereld Auto who informed me that he had been forwarded my enquiry from Van Kronenburg. Buddy is located pretty close to where I live (about a half hour drive) and can install and setup the KMS system on my Ghibli.
Above is a picture of the hardware which would replace the two OEM ECUs. He's given me a quote for fitting the system and I'm going to see if I can integrate the PLX R-500 into the installation to give me realtime feedback on the road. I've a feeling that the lambda probe that came with the R-500 is the same as that utilised by the KMS (a Bosch wideband probe) and I'll also see if I can integrate the Exhaust Gas Temperature (EGT) probe into the KMS, with maybe some feedback to prevent the exhaust gasses getting too hot and damaging the turbos.
The KMS can also control boost pressure, although I'm uncertain if it can tie into the TurboSmart solenoid valve. This would unfortunately make my E-Boost redundant, but I think the KMS would offer greater control over boost pressure with the added advantage of being able to tie in boost pressure maps with fueling and engine load.
I'm waiting to hear back from Buddy with answers to a couple more questions that I have, but if the price is right then I think I might go along with this option. The KMS is certainly much cheaper than MoTec and I like the idea of it being local, since if it stops working then I can go 'round the manufacturer personally to get them to sort it out!
Also, Buddy appears to have worked on a number of Maseratis, including a Ghibli fitted with a V8 engine from a Quatroporte in which the owner wanted the KMS system fitting. He also told me something interesting about the intercoolers from a Golf Mk4 TDi...
The Magnetti Marelli system installed by Maserati is very old compared with today's technology. To put it into context, this system was installed in Ford Sierra Cosworth cars back in the late eighties and early ninetees. The clock speed for this system is around 4MHz. Compare this to a modern laptop or desktop computer, which runs with a clock speed an order of magnitude quicker at 4GHz. In fact, you'd be luck to find a microprocessor that runs less than a hundred times quicker than that utilised in the OEM engine management system.
The effect of this is that the engine management system currently installed in my Ghibli makes significantly less decisions per second than the systems installed on typical modern street cars. Things happen and change very quickly in a car engine, particularly when it's reving at 7000rpm. That's over 100 revolutions of the crank every second, which is 25 detonations per second per cylinder.
The way an engine management system works is by taking a snapshot of the engine, which means recording the values of certain engine parameters at a moment in time by reading the output signals being sent from the various sensors located around the engine. The key sensors on the Ghibli are the inlet air temperature and the inlet air pressure for example. In order for the engine management to make decisions, it needs some reference data. This reference data comes in the form of look-up tables, which is just how people used to make complicated calculations in the days before electronic calculators. These are called maps.
A 2 dimensional map allows you to look up an answer if you know two numbers. Let's say that the inlet temperature is listed across the top row of the look-up table and the inlet pressure is listed along the left column of the table, the answer can be found by drawing a line down from the top inlet temperature value and a line across from the left inlet pressure value. Where the two lines cross is where the answer lies.
The answer tells the engine management system to perform a certain task. However, this answer is only applicable to the moment in time when the two input values were recorded, so if the system is too slow, by the time it carries out the response, it will no longer be applicable... a delayed response.
If there are more sensors, then the management system needs more look-up tables in order to know what decision to make, which takes more time. Also, the decision may require more than one response... for example the amount of fuel that is required and the time at which to ignite the fuel may need to be adjusted. Each additional response takes additional time.
My point here is that the quicker the management system can make decisions and respond, the more effective it is. In the Ghibli, it's obvious that the speed of the engine management system is not optimal, simply by virtue of the fact that TWO management systems are required, one to control the amount of fuel being injected and one to control when the fuel is ignited. So even when the Ghibli was being manufactured, it seems to have been recognised that processing speed was a limiting factor. Therefore, increasing processing speed should have a significant effect on the performance of the engine.
The next issue with the OEM management system is its lack of diagnostic capability. The only real way of determining if the system is making the right decisions is by measuring the performance of the engine, either by having a feel for the car or by sitting it on a dyno. The system does not have an easy way of monitoring what decisions are being made, except by taking the car to a specialist who can tap into the system and monitor what it is doing, which requires a degree of skill, technical knowledge and the appropriate equipment.
Modern management systems however offer a convenient way of monitoring what is happening. They generally have a convenient port into which you can plug a computer and with special software, you can record and review what decisions are being made.
Finally, if you recognise that sub-optimal decisions are being made, you need to be able to change the decision making process, which generally means altering the look-up tables or maps. Again, on the OEM management system, this is not a trivial thing to do and requires removal of the system and some special hardware to reprogramme the memory chips on which the data is stored. Also, the way this data is stored is not very user friendly, as it is stored in a way that is convenient for the microprocessor to read and not the car owner.
The beauty of a modern system is that all these things are possible. With modern software, you programme the data in a form that is easy to understand (using decimal numbers instead of hexidecimal or even binary code required for older systems) using a laptop computer. You can also record data from the engine, review it at a leisurely pace, determine what changes you want to make, update the look-up tables and see straight away the results... without having to go to a chip-tuner!
So, after contacting Van Kronenburg and discovering that they're a Dutch firm (much to my delight, since it means that they're local), I was a bit downheartened that I never heard back from them and it is then that I ordered the PLX Devices R-500 datalogger. However, I subsequently received an email from Buddy at Wereld Auto who informed me that he had been forwarded my enquiry from Van Kronenburg. Buddy is located pretty close to where I live (about a half hour drive) and can install and setup the KMS system on my Ghibli.
Above is a picture of the hardware which would replace the two OEM ECUs. He's given me a quote for fitting the system and I'm going to see if I can integrate the PLX R-500 into the installation to give me realtime feedback on the road. I've a feeling that the lambda probe that came with the R-500 is the same as that utilised by the KMS (a Bosch wideband probe) and I'll also see if I can integrate the Exhaust Gas Temperature (EGT) probe into the KMS, with maybe some feedback to prevent the exhaust gasses getting too hot and damaging the turbos.
The KMS can also control boost pressure, although I'm uncertain if it can tie into the TurboSmart solenoid valve. This would unfortunately make my E-Boost redundant, but I think the KMS would offer greater control over boost pressure with the added advantage of being able to tie in boost pressure maps with fueling and engine load.
I'm waiting to hear back from Buddy with answers to a couple more questions that I have, but if the price is right then I think I might go along with this option. The KMS is certainly much cheaper than MoTec and I like the idea of it being local, since if it stops working then I can go 'round the manufacturer personally to get them to sort it out!
Also, Buddy appears to have worked on a number of Maseratis, including a Ghibli fitted with a V8 engine from a Quatroporte in which the owner wanted the KMS system fitting. He also told me something interesting about the intercoolers from a Golf Mk4 TDi...
Ghibli prices on the up
I received an email recently from a Ghibli owner in France who had sold his MY93 car and was looking to purchase another Ghibli here in the Netherlands. He informed me that Ghibli's are relatively uncommon in France and as a result, the trend in prices had bottomed out and were now on the up.This is probably spurred on by the recent introduction of the new Maserati GranTurismo and the ongoing success of the Quatroporte raising the profile and perceived value of the Maserati brand. The result is that more people are becoming aware of the Maserati brand and are now considering Maserati as a viable option. Unable to afford the high prices of the new cars, people start to look at second hand purchases, increasing market demand and hiking up prices.
So I just had a quick look on AutoTrader's Dutch website and indeed, although you can pick up an older Ghibli for a pretty low price (this red MY93 Ghibli (with interesting graphite coloured wheels that have obviously been painted) can be picked up for about €11,000), general prices seem to have remained at least stationary since last year and for more recent versions of the car they do indeed seem to be increasing (this dark blue MY99 Ghibli is listed at €22,000)!It would therefore appear that now is a good time to purchase a Ghibli and that the car is set to become a classic!
Monday, May 07, 2007
Boost contoller sensitivity
The last few drives I've been playing with the boost controller sensitivity setting. It was set to 20 (not that this means anything since it's application specific), so I tried increasing it to 30 (although I was a bit weary since this is quite a big step).
I drove to Amsterdam on Saturday evening so got a good opportunity to load up the engine on full boost on the A4, keeping a close eye on the TomTom for cameras. I noticed that sustaining full boost gave the tell tale signs of a too high sensitivity setting, since the pressure gauge needle "wobbled". Before switching off the engine when I arrived at the first port of call, I quickly checked the maximum boost attained during my journey (my boost controller records the highest achieved boost level) and noticed it was high at 1.26 bar.
On one occassion during the drive I noticed a distinct hesitation during accelleration at maximum boost. It lasted a fraction of a second and I'm pretty sure that it was the ECU cutting the fuel supply to the engine after detecting a condition that it considered a bit on the extreme side.
While driving, the boost seemed to be responding very quickly to the throttle, which was nice. However, the wobble at full boost and the pressure spike are not really sustainable if I want longevity from my engine. Therefore, I reduced the sensitivity down to 25 before my next drive. This time, I no longer observed the wobbling needle at full boost, although when I checked the maximum attained boost level at the end of the drive it was still a bit high, indicating that the car was over boosting, I think in the lower gears where the revs ramp up quickly. I've therefore trimmed a little more off the sensitivity to prevent the spike a little, down to a setting of 23.
I very much doubt that this will have an impact on my ultimate power rating from the car, but soon I will be able to get a better idea since my PLX Device R-500 data logger arrived today! I'm not exactly sure when I'll get to install it and will probably have to do so over a few weeks (after first buying a solderer... I've not done any soldering since I was at school so I'm a little anxious about tapping into the car's wiring). I'm looking forward to seeing the data though... should give quite an insight into what's happening with the car...
I drove to Amsterdam on Saturday evening so got a good opportunity to load up the engine on full boost on the A4, keeping a close eye on the TomTom for cameras. I noticed that sustaining full boost gave the tell tale signs of a too high sensitivity setting, since the pressure gauge needle "wobbled". Before switching off the engine when I arrived at the first port of call, I quickly checked the maximum boost attained during my journey (my boost controller records the highest achieved boost level) and noticed it was high at 1.26 bar.
On one occassion during the drive I noticed a distinct hesitation during accelleration at maximum boost. It lasted a fraction of a second and I'm pretty sure that it was the ECU cutting the fuel supply to the engine after detecting a condition that it considered a bit on the extreme side.
While driving, the boost seemed to be responding very quickly to the throttle, which was nice. However, the wobble at full boost and the pressure spike are not really sustainable if I want longevity from my engine. Therefore, I reduced the sensitivity down to 25 before my next drive. This time, I no longer observed the wobbling needle at full boost, although when I checked the maximum attained boost level at the end of the drive it was still a bit high, indicating that the car was over boosting, I think in the lower gears where the revs ramp up quickly. I've therefore trimmed a little more off the sensitivity to prevent the spike a little, down to a setting of 23.
I very much doubt that this will have an impact on my ultimate power rating from the car, but soon I will be able to get a better idea since my PLX Device R-500 data logger arrived today! I'm not exactly sure when I'll get to install it and will probably have to do so over a few weeks (after first buying a solderer... I've not done any soldering since I was at school so I'm a little anxious about tapping into the car's wiring). I'm looking forward to seeing the data though... should give quite an insight into what's happening with the car...
Subscribe to:
Posts (Atom)