Diagnostics Interface
 

Is there a wiring diagram for the FSE/FSR on here?
Where is the diagnostics / trim port and does anyone know what protocol it uses?

I need to raise my idle a little.

Not sure where the connector is located on the FSE/FSR, but what I am sure about is that you will need the Technoresearch VDST software and cable or equivalent. The cable contains a level shifter to interface with the Magnetti Marelli ECU, it is not a straight RS232 interface. Also, on the version I have for the Ducati, I cannot adjust idle speed, just mixture trim, TPS, and diagnostics. The one for the FSRs ECU may be different thouigh.

Most ECU's use a standard protocol although they have varying connectors and pin outs. Many Magneti units use a "K line" protocol which is either KWP2000 or ISO something or other. That means there is a fair chance they can use an interface such as the ELM327 which is for OBD2 to USB. This device will operate with most recent standard protocols such as those mentioned and many others provided you can connect the correct pins. Unfortunately Gas Gas do not give a pin out for the ECU in terms of fully labelling what the pins are. If I get to the bottom of this and get it working I'll let everyone know as it'll be a lot cheaper than the VSDT which is probably the same thing packaged for a specific bike.

Because of this nonsense, which is used to keep things proprietary, European and I believe US legislation has been enacted to force vehicle vendors to use standard protocols and standard connectors and pin outs. Unfortunately it only effects the very latest released vehicles, not sure about bikes, and manufacturers are of course finding loopholes in law to continue their restrictive practices. One day we might be able to buy a cheap tool from the local car parts shop and work with any vehicle.

I have dissasembled the interface and drawn out the circuit. The Tx line from the ECU uses part of a standard Maxim 232 Rs232 level convertor. The ECU side of the level converter has an extra pull up resistor giving the line in from the ECU a nominal voltage of +6v. This is with respect to ground which is common to both the ECU and the PC it is connected to. This part is most likely standard TTL to rs232 conversion, ie inverted and voltage converted. The Rx or input line to the ECU is slightly different using a single transistor with a diode clamp. I would guess that in the ECU there is some kind of differential bus tranceiver to either a RS422 or 485 microcontroller interface. As for the communication protocol used it is probably home grown or using a subset of one of the known ones.
With regard to your original idling problem try playing with the throttle body bleed past screw which is the one visible through the plastic cover over the throttle cable on the kickstand side.
The service port is a 3 wire connector located under the seat just behind the tank. Probably identifyable by a rubber blanking plug in it to keep dirt and moisture out.
Hope this helps

I assume your talking about the Technoresearch inteface cable dongle. Yeah, I would expect it to be a TTL (ECU side) to RS232 (PC side) translator. Maybe I'll pop mine open and see if its the same. (IAW-5AM ECU). I know the Ducati uses a CAN bus but I don't beleive the comm port is part of it.

Mine is the older version dubbed MDST. It is Technoresearch's but plastered with Magneti Marelli branding. Even the software licence is with them. Its not Xp or later compatible and the serial lead is just that. The "dongle" which is purely for software protection sits in the parallel port which a lot of modern PCs do not have. I think the more modern one which Nick showed me had the dongle in a usb port.

Do you know which of the pins of the service port is output?
I was planning on trying some standard diag s/w to see if it can interpret the signal but on the only wiring diag I can find (2005 User Manual) it only identifies earth as the middle pin. The outer 2 pins aren't named.

As far as I can find the ECU is a Magneti Marelli IAW 15P not that this helps much as there seems no information on this to be found using Google. A lot of Magneti units use K line (KWP or ISO) protocol so standard s/w might work to read it. Not sure about writing. I'll try the bleed screw but I'd like to be able to read and adjust the ECU myself. Mainly to experiment with getting it to start easier from cold but the initial issue is stalling. The idle was set at 1800 in summer by the dealer but with the cooler air now it's dropped and will stall sometimes when idling. Maybe the IAT's gone - I should check that but it'd be nice to see everything on a screen rather than poking around with a multimeter.

This banter is way way above my head but I thought you guys, especially Glenn, might be interested to know that my local tuners at Silverback Performance are now able to map the Magnetti Marelli U59 ECU that Ducati uses. No longer limited to only trim, tps etc.

The output pin from the ECU is numbered pin 3 on the connector. On my bike the wire colour is blue.
Tecnoresearch had the Gas Gas ECU in their Direct Link Race Performance product but when I enquired with them they at first denied it, then said it would not be supported if I bought it and now references to the GG ECU appear to have been removed from their website. My guess is it was a licencing or copyright issue. I am sure the GG ECU is re-mapable by the end user just like all the other Marelli ECUs. The diagnostic tool , apart from making sure all your analogue to digital ducks are in a row (sensors), only allows adjustment of 1 parameter which is disappointing. I dont think it is unrealistic to dump the rest of the eeprom and figure out what the numbers mean or to change them and se what happens!

Ross,

Yeah, I read about that on ducati.ms. The Hyper/1098 ECU (IAW-5AM) as well. I suspect its a matter of knowing how to access/unlock the different areas of memory in the device. I imagine this was a closely guarded secret at Ducati being they like to sell DP ECUs for big $$. My brother and I are working on his Buell ECU right now, tweaking a race map for use with a wideband O2 sensor/controller. This stuff is fun. Back when I had my Cannondale I was able to hack the map data structure and edit the maps with a simple hex editor instead of the $1200 software.

I've just come across a product called TFI from Dobeck or Techlusion. It sits between the injector and ECU and allows you to amend the fuel delivery as you would with a carb system. This looks a better alternative to messing with trim and idle speed on the main ECU.
http://www.dobeckperformance.com/

This functionality should be built in as standard for competition bikes.

I had one on my HD VRod. All it is is a pulse stretcher device, extending the injector pulse width based on some small potentiometer adjustments. There is a newer all digital version as well. They work OK, but are no where near as precise as changing the map. The adjustments are limited to three RPM ranges, so you wind up being rich in adjacent areas of the map to the area you are trying to correct. Also, you can only increase pulse width (add fuel).

What is your problem? If its an idle issue you will be much better off adjusting the trim and air bleed. For what you pay for the TFI you can have the Technoresearch VDST, and do your trim, TPS reset, and diagnostics. If the base map is OK then the TFI is a waste and just adds complexity. The TFI is mainly used as a quick way for adding fuel when an aftermarket exhaust is used on a modern street bike, as they are set lean anyway from the factory for emmisions reasons.

I want to adjust the idle and I want it to start easier.

Then get the VDST, adjust trim and air bleed screw. The TFI is useless for this.

The dealer who has quite a bit of experience with GG and also Ducati (who use Magneti) couldn't get it any better. Said there was limited scope on the GG ECU. The TFI expands the adjustment quite a bit, even if only richer. Also someone else I know had similar problems on a Quad and it couldn't get sorted until a TFI was added. He reports it now starts easy in any condition and even pulled a bit better, not that I have a problem with that.

Not saying I'm getting a TFI but it sounds attractive from where I'm sitting. Buying the VSDTS could be an expensive test. Why would I do better than the dealer?

If the problem is a lean area of the map, the TFI can help. I cannot see how it could improve starting, because it is bypassed and not functioning when the bike first starts. It takes a few seconds to boot before it does its thing. This is because the ECU is applying a heavy start enrichment factor of its own at his point. There is a yellow LED that flashes while this is happening. Once this is complete, it will enrich the idle if the low speed range adjustment is set above zero. I'm sure mine worked this way, but if by chance Dobeck made a design change maybe you should call and find out.

The ECU trim parameter is fixed pulse width change, that is it adds or subtracts (on my IAW-5AM) approx 13us for each number change. It is not a % change of width, thats why its effect is primarily at idle and low demand areas of the map, where the pulse width is more narrow (leaner).

In the USA, the VDSTS, or VDST Standard (for one ECU family) is $200. A TFI is around $250+.

I assume your TPS was checked/reset and no codes are present. I'm by no means an experienced FSE guy, but I'm comfy with EFI systrems. I don't recall anyone here having to resort to adding a TFI to get the bike to run right, once the race map was enabled in the ECU.

You may be able to do better with the VDST software than the dealer because you wont be paying by the hour and will have the abilty (albeit limited) to play around with the fuel load setting and see if you can get a solution that suits you. You will also be able to spend time checking the interpreted sensor readings against actual conditions to see if there are any anomalies there.

What vintage is the FSE? I am sure its on here somewhere but I cant find it.
What idle rpm do you get and what were you looking to get?

GMP: Why can't the TFI work immediately? It'll boot when you turn the power on. It's not running Windows, it should be initialised as fast as the ECU! Maybe it doesn't work straight away for some reason but there's no reason it couldn't work from 0 revs.

The advantage over VSDTS trim that the TFI appears to have is there are separate settings for idle, acceleration and wide open.

Incidentally the bike runs fine but it doesn't like starting. From cold it will crank forever on a closed throttle with absolutely no sign of life. I have to play a game where I open the throttle, start cranking, let the throttle close and if it catches keep it cranking for a while or it'll die again. On the kickstart I have only once managed to start it and it took a lot of kicking leaning against a tree. I'm assuming it's too lean to start but that is an assumption not proven fact.

It's a 2006 model by the way and I take the point about having more time than the dealer.
I took it to him and asked for the valve clearances to be checked and the fuel trimmed. He did that and gave me the settings before and after having richened it a little but admitted it wouldn't start any easier. The only noticeable difference is it takes longer to warm up.

Read my post again.

How the ECU works during starting is to continue to increase the enrichment factor (fuel) while cranking if the bike does not fire. I would look into the throttle angle, air bleed, TPS, and check the codes. If the ECU is one that allows trim adjustment, and it is, (open loop, non-emmisions) you should be able to trim it up and down to the point of it stalling from being too rich/lean. I have done this on my Ducati with the open loop race ECU.

One thing I forgot, does that system have an idle air compensation stepper motor valve? I'm too lazy now to look it up myself.:D

I have and it's still not clear! Are you stating fact that the TFI does not change the pulse at startup or making an assumption? There is no reason why the TFI could not operate at 0 revs but I have no idea if it actually does. Let's put that question on the shelf unless we have a definitive answer from measuring or TFI themselves.

Well if it is supposed to keep enrichening until it fires something is broke on mine and I suspect many others. There is no air compensation stepper on the circuit diagram. What do you suspect - not enough air?

I'm going to try connecting a car battery to test if it spins any faster - was also wondering if the little battery can't supply enough cranking amps. It does seem quite slow turning.

How much clearer do I have to get?

I HAD ONE ON A VROD. IT WORKED THAT WAY. I DIDN'T MAKE THIS UP. Call Dobek or waste your money, I don't care.

The battery voltage being low is possibility.

Do a little research on how EFI systems function, understanding this this will help your troubleshooting immensly.

I'm finnished with this thread.

So I question how you are sure about something and you get angry? I take it you measured the pulse output in the first couple of revolutions and compared it to the input then? No you didn't did you? You're just guessing. I'm waiting for an answer from Dobek.

Update:
Sounds like they start at 800 rpm from Dobek

With my 06 in cold weather you have to reset the ECU at least once or you will crank all day or until the battery is flat (not long with this setup). For some reason the IAT does not read first time or doesnt stabilise quickly enough.

A typical cold morning start would be...crank a couple of times....turn off...back on....crank again and it starts. Its been like that since I got it.
As soon as it starts it will idle properly and is ready to ride off down the trail. No warmup, no bogging and certainly no chance of stalling.
After that it starts first touch everytime until it has been standing for a day or so.

This procedure is mentioned (well sort of) in the user manual which is a collection of Photos of older models, sentences with missing words and a piss poor spanish translated component names so it is easy to miss.
Hope that helps

Not sure it helps but makes me feel better that mine is "normal".

The only difference is on mine I usually have to open the throttle and close it while cranking for it to catch. What's more annoying is it's reluctance to fire up easily in gear when warm. Had a few times stalled in the middle of a race having to find neutral rather than press and go. However I've fitted a Rekluse which should stop that problem by preventing stalls once I've got it setup properly. Of course I absolutely must find neutral to start with that fitted if I need a little throttle to start it!