Transient Throttle Delta Load - stuck at 0.6% !!!

Questions about the Platinum Pro Harness 350Z and R35? Ask here.
HaltechMatt
Haltech Staff
Posts: 130
Joined: Tue Sep 25, 2007 6:39 am

Re: Transient Throttle Delta Load - stuck at 0.6% !!!

Postby HaltechMatt » Tue Jul 14, 2009 2:15 pm

and from the online help of 1.05

Transient Throttle Enhance

Transient Throttle Enhance function aids in improving engine response when the throttle on a conventional fuel injected vehicle is opened rapidly. The Transient Throttle Enahncement works by adding additional injection pulses (asynchronous enrichment) and enriching the current fuelling pulses (synchronous enrichment) when it sees a change in throttle position. The setup allows the tuning of:

* How much enrichment is needed for a given RPM, start throttle position and TPS rate of change
* The biasing of the enrichment between asynchronous and synchronous types according to a given RPM
* The speed at which the synchronous enrichment decays

Note the rate of change of throttle position is how far the throttle moves over a 10ms time frame. A fast throttle movement could have a TPS rate of change of 15% this means that in a time frame of 10ms the throttle position moved 15%. A slow throttle movement could have a TPS rate of change of 1%, which means that in a time frame of 10ms the throttle position only moved 1%.
Settings

* Number of Async Pulses - This is the maximum number of extra injector pulses that can happen per injector output during an enrichment event.
* Async Delay Time - The minimum time between asynchronous enrichment events. This is to stop excess fuelling when constantly pumping the throttle.
* Delta Dead Band - To prevent excessive enrichment due to slight throttle jitter, only a rate of change of throttle position greater than this value will trigger an enrichment event.
* Transient Throttle Trim - The trim value scales the fuel enrichment. 0% makes no change to the fuel enrichment. 100% will add an extra 100% to the enrichment. Setting this to -50% will remove 50% from the enrichment.

Transient Throttle Tables
Enrichment Sensitivity

This map allows for the tuning of the amount of enrichment needed according to a given RPM and starting throttle position. In general more enrichment is needed at a low throttle position because a change will cause a larger in rush of air than at a higher throttle position. At a low RPM more enrichment is needed because of low air speeds and the atomisation of fuel is poor. This becomes less of a problem as the engine goes up in RPM and airflow increases.

The value in this table represents the enrichment needed at the maximum Delta Load. The maximum Delta Load is the largest change in throttle that you can make in the sample period of 10mS. In simple terms, this is the amount of enrichment for a full throttle opening as fast as you think you will ever open the throttle. Display the Transient Throttle Delta Load channel on a display or gauge to determine the magnitude of maximum Delta Load that you can achieve.

This table is the first to be tuned but also make sure that the right hand column of the Percentage Enrichment table equates to the maximum Delta Load that you have selected.

Percentage Asynchronous

This table determines what percentage of the fuel enrichment is delivered as asynchronous enrichment. Setting 0% in the table means that all of the fuel enrichment will be delivered synchronously (ie no additional fuel pulses will be added, all enrichment will be delivered by increasing the regular injection pulse). Setting 100% in the table means that all of the fuel enrichment will be delivered asynchronously (ie the normal injection pulse remains the same and all the enrichment fuel is added through extra asynchronous pulses between the main injection pulses).

Note that if the enrichment value exceeds 9.5ms (the maximum size of an asynchronous pulse) the remaining enrichment will be delivered synchronously.

As a guide, at a low RPM, when the injection events are short and further apart, the asynchronous fuel enrichment helps deliver the extra fuel where its needed. At a high RPM asynchronous enrichment is not needed because regular injection events are more frequent.
Accel Ignition Adjust

This table allows the ignition timing to be adjusted during transient throttle events. The correction is an adjustment on the overall ignition timing value.
Percentage Enrichment

The Percentage Enrichment table allows you to proportion the amount of enrichment depending on how much change in throttle (Delta Load) that you make at any transient event.

The Enrichment Sensitivity table above is tuned for a maximum Delta Load event. To accomodate lower Delta Load events, the Percentage Enrichment table is applied.

The right hand column of the table should contain the largest Delta Load value in its axis value and 100% in its map value. This value should have already been determined when setting up the Enrichment Sensitivity table above.

E.g. If you anticipate that the largest throttle change that you can make in the sample period of 10mS is 25%, then the axis value should be set to 25%. The Enrichment Sensitivity table values are based also on this largest Delta Load case, so in the right most column value, you should set 100%.

For slower throttle movements (lower Delta Load values), less enrichment is required, so values less than 100% should be set in these columns.

Enrichment Decay Rate

This map determines how fast the synchronous enrichment (i.e. the enrichment added by extending the normal injection pulse) decays back to zero (no additional enrichment). The units are milliseconds per engine cycle.

E.g. a value of 0.5 will mean that the enrichment will decay 0.5ms every engine cycle. The bigger the value the faster it will decay.
Accel Ignition Decay Rate

The rate in degrees per engine cycle that the ignition offset will decay. The larger the number, the quicker the decay.
Coolant Temp Corr

This map applies a correction factor to the fuel enrichment based on the coolant temperature. E.g. 100% gives 100% extra fuel enrichment.

User avatar
BRIAN MP5T
Posts: 1106
Joined: Sun Feb 01, 2009 4:05 pm
Location: OTTAWA
Contact:

Re: Transient Throttle Delta Load - stuck at 0.6% !!!

Postby BRIAN MP5T » Wed Jul 15, 2009 4:29 am

Awesome, I think I will have to read it three times to get it, but thanks very much.

MR RIZK
Posts: 34
Joined: Wed Mar 04, 2009 12:37 am

Re: Transient Throttle Delta Load - stuck at 0.6% !!!

Postby MR RIZK » Wed Jul 15, 2009 7:23 pm

I read it 3 times, even tried reading only every second word :( I wish I was a tuner..

rcdash
Posts: 137
Joined: Fri May 22, 2009 11:55 pm

Re: Transient Throttle Delta Load - stuck at 0.6% !!!

Postby rcdash » Fri Jul 17, 2009 12:20 am

Thank you Matt for the data.

I was aware that delta load represented throttle change over 10 ms, but the problem with this implementation is that even with no throttle change, there is delta load jitter (hence a "dead zone" is required to preclude constant ad hoc fueling). Therefore it is difficult to separate the signal from the noise. With the 350Z firmware, perhaps we don't have control (as tuners) to some of the parameters that be helpful to us for transient throttle tuning (vs the Sport series for which that help documentation is accurate - on the 350Z, WE CANNOT ACCESS SEVERAL OF THOSE PARAMETERS!!!)?

How can anyone be expected to tune transient throttle utilizing delta load effectively when the signal to noise ratio is so poor?

Can we alter delta load to represent change in throttle over 50ms or 100ms instead (to increase the signal to noise ratio)? Or can the TPS value be averaged/smoothed in such a way to decrease the noise if the 10 ms period is fixed?

How is the length of the async pulse determined? I've noticed it sometimes lasts up to .5 seconds (logging transient throttle - async enrichment fueling).

Also, any insight on what the transient throttle-delta load (versus just delta load) represents?

Thanks again,
Raj

PS. I guess the bottom line is, thanks for the data, but how do we fix this problem???

HaltechMatt
Haltech Staff
Posts: 130
Joined: Tue Sep 25, 2007 6:39 am

Re: Transient Throttle Delta Load - stuck at 0.6% !!!

Postby HaltechMatt » Thu Jul 23, 2009 12:30 pm

I see how it is, yesterday all you want is information on how the system works so you can make informed decisions on what to modify - today you want to know how to fix your problems....Do you happen to know my wife....she employs this same tactic, today its "honey i think the car needs air in the tires" tomorrow she is signing the papers on a new BMW. At least she has good taste in cars.

OK ive got all my jokes out in 3 lines so i will get on to answering your questions ! (btw if i just offended anyone sorry - just trying to lighten things up)

Perhaps to clarify my past two (excessively long - if anyone actually read every word of each post without needing a beer at the end i would be surprised - posts) throttle pumps work like this:

- Rapid movement of throttle triggers throtle enrichment
- All injectors immediatly pulse 5 pulses in 10ms internvals for the legnth determined by the asynchronous enrichment map (more on how this is determined later)
- Each injectors "normal" injector pulse legnth is increased by the legnth determined by the synchronous enrichment
- The next engine cycle each injector pulse legnth is increased by the legnth determined by the synchronous enrichment less the enrichment decay rate

An example with numbers:

Base fuel delivery is determined to be 2ms
Engine is at 3000 RPM

Throttle is stabbed and movement is determined to be 10% movement of throttle over 10ms - this gives delta load of 10%
ECU looks at enrich sensitivity map and locates the 3000RPM, 10% delta load value - which is mapped at say 8ms (as an example)
ECU looks at % async enrichment for 3000RPM - which is mapped at say 50% (as an example)
All injectors pulse 5 times at 4ms per pulse in 10ms intervals (4ms is 50% of 8ms)
Each of the next "normal" injection pulses for 1 engine cycle is extended from 2ms to 6ms (2ms base fuel delivery, plus 4ms synchronous enrichment)
ECU looks for the Sychronous decay rate - which is programmed at say 1ms (as an example)
Each of the "normal" injection pulses on the second engine cycle after the initial throttle movement is 5ms (2ms base fuel delivery, plus 4ms sync enrich, minus 1ms decay = 5ms)
This trend continues until there is no further enrichment to the base fuel pulse or the throttle is stabbed again.

Again i understand the issue your having is that the throttle movement you are talking of is not fast enough to trigger the delta load value above the deadband threshold. This unfortunately is a limitation of the signal input to the ECU, being able to modify the 10ms time window is something that i have talked to the R&D guys about and they will test that out and see if it helps.

In the mean time, i have a couple of questions to ask in regard to driveability with this issue - do you notice any stuble or hesitation or is it just in the logs you notice a lean out? I have been frantically trying to get some info from OEM factory vehicles to help solve this issue for you guys and what ive found (which may be limited to the number of vehicles ive logged) but im having no issue (or at least less) on vehicles that are factory run with an airflow meter....vehicles using factory MAP sensors (yes there are still a few out there) see a lean out (logged AFR but no drivable difference). Perhaps there is something inherit here?

I dont have all the answers (dont tell my wife i just admitted that) but the more info we get from you guys the better we can make things in the future.

rcdash
Posts: 137
Joined: Fri May 22, 2009 11:55 pm

Re: Transient Throttle Delta Load - stuck at 0.6% !!!

Postby rcdash » Fri Jul 24, 2009 1:14 am

Thank you Matt. This is very helpful. (And I hope your wife doesn't read the Haltech forums! :))

I will use this and try to modify transient throttle tuning armed with this data. I will post up logs to document findings.

I do think that either the 10 ms window needs to be configurable (lengthened) OR the delta load signal needs to be filtered in a way that mitigates jitter (e.g. running average). The former suggestion seems like it would be easy.

In regards to whether you can feel the hesitation - yes. But, not always. I found that I was changing my driving style to avoid the stutter (no slight throttle movements). I'd like to not have to do that. Sometimes I think I would get it right, only to have problems under different conditions (rpms, load, etc). I felt like I was tuning transient throttle "in the dark" so gave up trying to get it perfect. Will give it another shot now that I know that async is giving 5 pulses every 10 ms... I must admit I wonder how you can coordinate async and sync pulses so that both occur without conflict (depending on rpm, etc). At 3000 rpm there should be about 4 sparks per second per cylinder. I'm trying to determine if the total fuel delivery is consistent regardless of where in the cycle any given cylinder is at the moment of throttle application. Perhaps these are details that are insignificant to what can be felt while driving however?

rcdash
Posts: 137
Joined: Fri May 22, 2009 11:55 pm

Re: Transient Throttle Delta Load - stuck at 0.6% !!!

Postby rcdash » Fri Jul 24, 2009 1:44 am

MR RIZK wrote:I read it 3 times, even tried reading only every second word :( I wish I was a tuner..


If you post a high res (5 ms) data log, we might be able to identify whether it is a rich or lean condition during shifts causing the issue. Upload snapshots to a place like photobucket, then link to it from a post here.

RyleyMA61
Posts: 29
Joined: Sun Jan 20, 2008 3:27 pm

Re: Transient Throttle Delta Load - stuck at 0.6% !!!

Postby RyleyMA61 » Fri May 21, 2010 12:07 am

I'm really glad I've found this thread, rcdash, I am having the exact same issues (very light throttle movement, lean out spike, sometimes feel it, other times just see AFRs shoot up) except with an E11v2 and Toyota 7M-GTE.

Did you find any way of fixing this? I am about to go out and fiddle with the percentage of async enrichment values for lower rpm (among a few other things) and see if it helps.

Thanks,
Ryley

rcdash
Posts: 137
Joined: Fri May 22, 2009 11:55 pm

Re: Transient Throttle Delta Load - stuck at 0.6% !!!

Postby rcdash » Tue May 25, 2010 6:32 am

Yes, it's mostly been tuned out. Once you know what's happening internally, it just takes some trial and error to get it smooth. You can use 5 ms data logging or just go by feel. Good luck.

RyleyMA61
Posts: 29
Joined: Sun Jan 20, 2008 3:27 pm

Re: Transient Throttle Delta Load - stuck at 0.6% !!!

Postby RyleyMA61 » Tue May 25, 2010 10:38 pm

and it was all in the throttle pump settings? can you say which bits you tweaked the most?

thanks mate

User avatar
HoSDo_earth
Posts: 179
Joined: Fri Jan 08, 2010 10:39 pm
Location: Udine - Italy
Contact:

Re: Transient Throttle Delta Load - stuck at 0.6% !!!

Postby HoSDo_earth » Sun Sep 04, 2011 3:48 am

Image

This is a Datalog from today morning: I'm experiencing no enhancement at all also if I set low Delta Dead Band (0,2 right now).

Speaking with some friend, seems like I'm suffering of drying out manifold while closing the throttle for long periods (i'm not using the cut off function). When I open the throttle again no difference between low or large variations AFR runs 19. It's not about the map i presume, in that area i'm usually running already too rich about 12. It's just that the throttle doesn't call the enhancement.
When you shift the gear and the needle on the tacho goes on red and scores 9000 rpm, that's bad...

User avatar
HoSDo_earth
Posts: 179
Joined: Fri Jan 08, 2010 10:39 pm
Location: Udine - Italy
Contact:

Re: Transient Throttle Delta Load - stuck at 0.6% !!!

Postby HoSDo_earth » Sun Sep 04, 2011 4:47 am

HaltechMatt wrote:This map allows for the tuning of the amount of enrichment needed according to a given RPM and starting throttle position. In general more enrichment is needed at a low throttle position because a change will cause a larger in rush of air than at a higher throttle position. At a low RPM more enrichment is needed because of low air speeds and the atomisation of fuel is poor. This becomes less of a problem as the engine goes up in RPM and airflow increases.

The value in this table represents the enrichment needed at the maximum Delta Load. The maximum Delta Load is the largest change in throttle that you can make in the sample period of 10mS. In simple terms, this is the amount of enrichment for a full throttle opening as fast as you think you will ever open the throttle. Display the Transient Throttle Delta Load channel on a display or gauge to determine the magnitude of maximum Delta Load that you can achieve.


Matt, I must ask you further explaination while I can't really understand the logic of this function anymore! First of all I'm on a PS1000 ECU just in case something would be different (but I don't think so)

The "Load Source" for the throttle pump is simply the rate of the throttle opening within 10 ms: testing the engine should reveal how low should this parameter be set to trigger the enhancement. However this part of the function is basically Throttle movement within time.

The "Delta Load" is the magnitude of the load request generated by the throttle opening. I assume (it is never stated) it is related to the "Load Source" of the engine so in my case it is basically the variation in MAP.

Now, the last thing that isn't clear to me is when you explain "the amount of enrichment needed according to a given RPM and starting throttle position": so the Sensitivity table is related to the starting position of the throttle meaning that the x axis it the actual Throttle position not the "Load Source" fraction on 10 ms.

So what I really don't understand is: how low should I start with the Delta Dead Band to look for the right enhancement? I tried some days ago a 0,1 setting but I immediatly got continuous rich mixture with AFR spike down to 10,2 while running at -67 kPa.
When you shift the gear and the needle on the tacho goes on red and scores 9000 rpm, that's bad...


Return to “Platinum Pro Harness”

Who is online

Users browsing this forum: No registered users and 1 guest