G4+ Fury Sequential shift cut

[VtrSp1]

I have a G4+ Fury installed in my race car with sequential gearbox.

The ecu is awesome and all work very good. I need only a few clarification about the Sequential Shift Parameters/setting .

As i mentioned before, i use the ecu to do the Shift cut for my Sequential Gearbox in Closed loop mode with Loadcell Knob and Gearposition sensor with Ignition cut % set at 90% and a Ignition Reatard at -15° for reintroducing torque strategy at 20 mS. In the Last weekend i tested them on track and when i do the Upshift strategy i have a big bang in the exhaust . it's normal because i cut the ignition at 90% and the rest 10% is retarded at -15°. I set the retard to -15° because i feel the shift lever smooth than a -10° retard.

I have try to set the Fuel trim shift cut to a negative value to reduce the fuel amount in the shift cut but it's not possible to insert a negative value.

How i can remove fuel in the shiftcut strategy to reduce the bang in the exhaust?

In the log i take a look at Gear Parameters and when the barrell move, the Gear number go to 0 for a little moment during shifting. My Gear tollerance is set to 0.1V because i want the shift cut quite safe.

If the Gear tollerance is low value, the power re-engage is safe for a full dog engage vs a high gear tollerance number?

 

[Adamw]

It sounds like you understand the systems well.

 

42 minutes ago, VtrSp1 said:

I have try to set the Fuel trim shift cut to a negative value to reduce the fuel amount in the shift cut but it's not possible to insert a negative value.

Try a +10% fuel trim.  I know this goes against what your logic would expect but I find this usually works better at minimising the "bangs".  On other ECU's I use that allow both positive or negative fuel trim I still find adding fuel is usually the best.

 

45 minutes ago, VtrSp1 said:

How i can remove fuel in the shiftcut strategy to reduce the bang in the exhaust?

If you really want to try a negative fuel trim, you can do it using a 4D fuel table.  Set the table activation to "Virtual aux 1".  Set up Virtual aux 1 to become true when ignition cut is >89% and <=90%.  The 4D fuel table will then only become active when the 90% gear cut is applied. 

 

51 minutes ago, VtrSp1 said:

In the log i take a look at Gear Parameters and when the barrell move, the Gear number go to 0 for a little moment during shifting. My Gear tollerance is set to 0.1V because i want the shift cut quite safe.

If the Gear tollerance is low value, the power re-engage is safe for a full dog engage vs a high gear tollerance number?

 

Yes, this is quite normal to see a short "gear 0" between each gear when using closed loop shift.  The gear tolerance affects the position that the ECU considers that the next gear is fully engaged.  My tolerance is usually more like 0.2v on the few closed loop setups I have done so your 0.1V maybe a little conservative.   This takes some experience to get right but I suggest logging most gear related parameters at 100Hz and try a couple of larger tolerance variations, you can usually get a pretty good idea by looking at gear lever force and barrel position.  If the tolerance is about right you will often see little dip in gear lever force at the end of the shift sequence while the barrel is still moving (this is the dogs getting pulled together by the tapers when torque is reintroduced). 

[VtrSp1]

Adamw, thanks a lot for the helpfull reply .

Very interesting the strategy for quite the exhaust bang with add fuel in the shift...normally i see the opposite, so cut the fuel at 100% for don't have extra fuel washing the exhaust with ignition retarded. I want to remove all bang possible due to better life of turbo and exhaust manifold.

I take a look at my gear lever force and i see the dip of force but if you can see and give me some consideration about that , this is very helpfull for me :-)

I have a internal log at 100 Hz of my track session , so i can upload and you can take a look at my gearshift parameters. I logged at 100 hz all the shift parameters.

 

[Adamw]

The dip you see in gear lever force is the start of the cut when the dogs let go. Often you will also see a little dip in gear lever force at the end also when the torque is reintroduced (but not always).

Your gear shift doesnt look too bad to me, you have on average about 120ms cut in the lower gears and about 80ms cut in the higher gears so this is not bad for most car gearboxes.  The volt tolerance looks about right to me as the logs show power is being re-introduced while the barrel is still moving.

From what I can see I dont think there are any big gains that can be made.  You may be able to make the shift feel smoother with some small tweaks to tolerance and upshift force, but I think in terms of performance you are not far off.  Does the gear shift feel ok or is there something you are not happy about (apart from the backfire)?

[VtrSp1]

Adamw, thanks a lot for the awesome explain !! Your explain is about same as my when i see the log after the race .

The cut time in high gear (3-4-5) is about 70-80 mS and it's not bat for me, it's the time that i expect from this type of gearbox with shift lever. I think it's possible to improve- reduce this time with modify the mechanical rotational mass (more light flywheel and racing clutch without with small diameter unsprung).

The logs show the power is re-introduced with the barrel is still moving,  normally this  can cause premature dog wear...

What i can do for have the re-introduced power with barrell stopped? I can play with a lower gear tollerance?

The lever fell smoother but not smoother than if i dab the clutch for the same shifting..

In the first part of lever pull i feel a hard shot ("tac ") with the Upshift and this isn't present when i dab the clutch or left the foot from the throttle to unload the dog...

I have the same issue a few month ago when i used for the Gear cut the Geartronics EasyShift Ecu (coil cut) in closed loop mode.

This strange feel in the lever i think in the log is showed by the firs peak of loadcell signal..

 

[Adamw]

Sorry for the slow reply.  

On 9/12/2019 at 9:57 PM, VtrSp1 said:

The logs show the power is re-introduced with the barrel is still moving,  normally this  can cause premature dog wear...

What i can do for have the re-introduced power with barrell stopped? I can play with a lower gear tollerance?

In my experience it is pretty normal to reintroduce power while the barrel is still moving the last little bit, and I havent seen any wear I thought was excessive. You are correct in that you would lower the voltage tolerance if you wanted only re-introduce power when fully engaged.

 

On 9/12/2019 at 9:57 PM, VtrSp1 said:

The lever fell smoother but not smoother than if i dab the clutch for the same shifting..

In the first part of lever pull i feel a hard shot ("tac ") with the Upshift and this isn't present when i dab the clutch or left the foot from the throttle to unload the dog...

What is the "start shift validation time" set at?  5-10ms is where I normally have it.  Also have you tried lower upshift force to see if it makes it feel better or worse?

[VtrSp1]

On 9/30/2019 at 9:22 AM, Adamw said:

Sorry for the slow reply.  

In my experience it is pretty normal to reintroduce power while the barrel is still moving the last little bit, and I havent seen any wear I thought was excessive. You are correct in that you would lower the voltage tolerance if you wanted only re-introduce power when fully engaged.

 

What is the "start shift validation time" set at?  5-10ms is where I normally have it.  Also have you tried lower upshift force to see if it makes it feel better or worse?

Adamw, don't worry for the slow reply :-)

My "start shift validation time" is set at 2 ms. My upshift gear lever force is set to 44 Nm , i check a few log and the cut start before the barrel move (i take a look at gear lever force and barrell mV value). I can try to low the upshift gear lever force a bit but i think i can have false trigger when in the track.

[mapper]

I'm using fuel cut most of the time to prevent the very hard bangs during shifting. But the downside is, the power cut feels a bit slower. 

I would really like if LINK will do a IGN And Fuel cut option.

[Adamw]

11 hours ago, mapper said:

I'm using fuel cut most of the time to prevent the very hard bangs during shifting. But the downside is, the power cut feels a bit slower. 

I would really like if LINK will do a IGN And Fuel cut option.

Whats your theory on why fuel and ign cut would be better?

There will always be much slower recovery if you cut fuel.  When you cut ignition only, since there is still fuel in the cylinder, as soon as you re-enable ignition there is torque available almost instantly.  In contrast, if you cut fuel, when you re-enable it the fuel only gets injected into the port on the next scheduled event for the next cylinder, it then has to sit in the port until the next intake stroke, then it gets compressed on the next stroke, and then finally combustion and torque is produced after that.  So depending on timing you are usually waiting 3 crankshaft revolutions until you get power again after a fuel cut.  Cutting ign and fuel together will have the same delay as cutting fuel only.

Have you tried a +ve fuel trim with ignition cut?  It seems to work well for most of the cars I work on.

[VtrSp1]

Adamw,

after some track test with the race car i have interesting data about the shift cut strategy..

All test was made in the same track with same car setting (tyre, power level etc)

-  Ignition cut at 90% , ignition retard -10° and no Extra fuel trim in the shift cut strategy.

   Pros: cut instant, fast and hard, no fuel lean after the cut end strategy , no gear re-bounce

   Cons: Hard bang in the exhaust that can destroy the turbo /manifold and also engine

- Ignition cut at 90% ignition retard - 10° and add a fuel trim +10% in the cut strategy

   Pros: cut instant , fast and hard , no gear re-bounce

  Cons : Hard bang in the exhaust and the 10% extra fuel trim don't solve the exhaust bang. The exhaust bang is same as without the +10% fuel trim

- Fuel cut at 90%, ignition retard -10° and No extra fuel trim in the shift cut strategy

  Pros : shi ft lever feel smooth , no exhaust bang

 Cons : cut not instant and fast as the ignition cut ( i use a low Gear lever force Nm for start the cut before), and if the shift  Lever travel isn't full and very precise i have a 2 time a gear re-bounce (shift from 3 to 4 and the box return in the 3 th gear). A lean situation after the gear cut strategy end

- Fuel cut at 80% , ignition retard -10° and no extra fuel trim in the shift cut strategy

 Pros: shift lever feel smooth than fuel cut 90% , no exhaust bang

 Cons: cut not instant  and fast as the ignition cut, 1 times a gear rebounce and lean situation after the cut strategy.

 

 

So for the cut power is better the ignition cut (it's fast and instant and i haven't gear re-bounce) but the downside is the bang in the exhaust.

Do you think with a low % of ignition cut, the exhaust bang is reduced?

The other test is reduce the ignition cut % to the number that can't give me a dog disengage for see what is the low % allowed for the dog disengage and after try to  use a little high % than the minimun finded but not too much?

If i use a 10% fuel trim in the shift cut strategy, this can help me in the lean situation after the fuel cut strategy for upshift?

 

[Adamw]

Sorry I have never used fuel cut much, only briefly experimented and found it to be way too slow.  With an ignition cut there is already fuel and compression waiting for you so as soon as you reintroduce spark you have instant torque.  With a fuel cut, once the shift is finished the ecu has to wait until the next scheduled injection event (maybe up to a full 720 deg later depending on timing), then after injecting fuel you have the intake stroke, then compression stroke, then finally torque another 360 degs later.

I would say the "lean spike" you see is just the oxygen going through the cut cylinders and is nothing to worry about.  I dont see any harm in adding a fuel trim if you want to give it a try.  You can also do some tricks such as use a virtual aux and a timer to add a fuel trim for a short time during/after the torque reintroduction phase only if you wish.

[mapper]

Sorry I've forgot about this thread. 

I got similary expierience as  Vtrsp1. Even with 15% fuel enrichment, i can not prevent big bangs (and following overboost) in all situation. So the only solution seams to be fuel cut. I'm using 60-85% cut depending on gear an up to 40 deg retard. 

Adam I was hopping that the cut is faster with the use of  ign+fuel cut, but now I've read your explaination, there wouldn't be any difference. 

That gear 0 is detected between shifts is a bit strange regarding gear dependant settings. As I mentioned above I use different cut leveles based on gear. Now if it is detecting zero for some time, it takes gear 0 cut mid shift. I think there should be abother status integrated "betweeen gears", so gear cut level of the gear shifted out is used for the whole shift process. 

[VtrSp1]

On 12/22/2019 at 11:06 AM, mapper said:

Sorry I've forgot about this thread. 

I got similary expierience as  Vtrsp1. Even with 15% fuel enrichment, i can not prevent big bangs (and following overboost) in all situation. So the only solution seams to be fuel cut. I'm using 60-85% cut depending on gear an up to 40 deg retard. 

Adam I was hopping that the cut is faster with the use of  ign+fuel cut, but now I've read your explaination, there wouldn't be any difference. 

That gear 0 is detected between shifts is a bit strange regarding gear dependant settings. As I mentioned above I use different cut leveles based on gear. Now if it is detecting zero for some time, it takes gear 0 cut mid shift. I think there should be abother status integrated "betweeen gears", so gear cut level of the gear shifted out is used for the whole shift process. 

Mapper, i have try with 20% fuel enrichment in one track lap and i can't prevent or soft the big bang in the exhaust. 

I'm about interested to your value about the fuel cut 60-85% and 40 deg retard...

In my experience, low gear need more fuel cut % than the high gear. The 40 deg retard is interesting, with only 15 degree retard i have a power reintroducing too soft and the engine bog...

In my car, at full boost the ignition is 15-16 ° Btdc, so if i retard more than 10 °and more than 20 mS time  , the power re-introducing is too slow and i feel a loss of power in the upshift.

The gear 0 detected between the shift is strange . I think the ecu use the strategy looked at the start of upshift and they don't change if during the upshift the value gear go to 0

[mapper]

I have to check in datalogs if the cut% stays the same even if 0bis detected durung shift. 

Settings really depend on power level and engine. I don't use 40drg retard on all cars. If 10deg works for your application, leave it. 

Back to fuel &ign cut. Shifting out of gear will be faster, if ign cimut is appkied instant, rather than waiting for next inj. event. 

 

 

[VtrSp1]

Mapper, i have check my log and the cut stay On when the gear detected is 0

[VtrSp1]

Adamw,

it's possible to set a minimun cut time when the cut end is set to barrel position?

This can help in a endurance race to minimise damage to the box if the gear position sensor or other fail..

So example..if i have the low upshift time at 85 mS, i want to set a value for minimun cut time to 70 mS

[Adamw]

10 hours ago, VtrSp1 said:

Adamw,

it's possible to set a minimun cut time when the cut end is set to barrel position?

This can help in a endurance race to minimise damage to the box if the gear position sensor or other fail..

So example..if i have the low upshift time at 85 mS, i want to set a value for minimun cut time to 70 mS

If the end shift mode is set to "gear barrel position" and your gear pos sensor fails, then in this case the "max shift time" setting takes over.  Typically you set your max shift time a bit longer than your normal longest shift - so if the sensor fails you will have a slightly slower shift than ideal but it will still cut and shouldnt be any harder on the gearbox.

[VtrSp1]

Adamw, I have an opposite situation..

the sensor report a signal Little low that is out of the volt tolerance so the ecu show me 0 gear but in real I’m on 3rd.. I pull the lever for upshift and the ecu start the cut from 0 but after 8 mS see 3rd gear and end the cut in 8mS..

 

so if I have a minimum shift time at 70 mS the power is not reintroducing early 

[VtrSp1]

This is what happen..

[Adamw]

Oh ok, I understand now.  I have not come across this situation before.  There is really no way in the current firmware take care of this, it will really need some new settings added.  Why is your sensor not detecting the gear properly? does it settle on a different voltage everytime?  Is there some mechanical backlash or something? 

[VtrSp1]

7 hours ago, Adamw said:

Oh ok, I understand now.  I have not come across this situation before.  There is really no way in the current firmware take care of this, it will really need some new settings added.  Why is your sensor not detecting the gear properly? does it settle on a different voltage everytime?  Is there some mechanical backlash or something? 

Adamw, 

Bad news for an setting update...:-( It's not possible to do it with a Virtuax Aux?For example activate a  rev limiter with cut when the gear lever force is over 60 Nm or volt and start a timer for the rev limiter cut as 60 mS?

My sensor not detecting the gear properly because i used a low volt tollerance 0.06V and the sensor signal often have a +/- 0.03 fluctation only in 3rd gear(Exampòe..the 3rd gear was set to 2.14 and often after the upshift o downshift i see 2.24 Volt that is out of volt tollerance), 4th looks perfect at +/-0.01 . This happens after a Box rebuild, before the signal was stable and perfect with 0.01 volt fluctation. 

I have check the +5V out to the sensor, the ground but all looks good. The magnet on my barrel is clean with no debris .  

I have try to rotate the sensor on the box cover to 20°,  i re-done the gear position setting and now after a few upshift and downhisft with no load to the wheels, the signal are more stable..

I attach a log and the calibration , the log show the few upshift and downshift in sequence 

new gear setting.llg new mods- boost control - gear setting.pclr

[VtrSp1]

Adamw, i have another question..

For double check in my log, i'm on planing to buy a new barrel position sensor with twin channel output..

Now i have the one channell version and it's configured as Gear position sensor..How i can configure the 2nd channel ?

[Adamw]

There is no functionality to take advantage of a dual track gear pos sensor at this time sorry.  

[VtrSp1]

@Adamw i need some a clarification about the sequential shift Volt Tollerance.

This thought has bored me for a while..

The example is in the screenshoot above.. my volt gear tollerance is set to 0.10 V , 3rd gear volt is set to 2.72V and 4th gear is set to 2.12 V

In the upshift from 3rd to 4th, in a closed loop situation, the ecu considered the next gear engaged (4th) when the voltage is 2.12 v + 0.10 V, so 2.22 V... I understand correctly?

If i watch all my log, the power is reintroduced very early ...in the screen shoot above at 2.47 V (logged at 100 mS in ecu internal logging)..

if i do the difference from 3rd to 4th is 2.72-2.12 V , so 0.60 V....If the power is reintroduced at 2.47 V   as the log show, the barrell and the dog ring is about at the half of full travel..(2.12 + 0.30 V = 2.42 V)

It's a ecu or software bug?

This is for all the gear and in all my log..

 

[Adamw]

3 hours ago, VtrSp1 said:

In the upshift from 3rd to 4th, in a closed loop situation, the ecu considered the next gear engaged (4th) when the voltage is 2.12 v + 0.10 V, so 2.22 V... I understand correctly?

Correct.

Can you attach the log so I can take a look.  I suspect it is just interpolation and missing data points but I would like to look for some clues myself.