Charge Dilution and PV

[joe_lyons]

So Joe, chiming here, I read my map load normalisation which I understand to be similar to bias global. At 1000 rpm my ratio is 1.02 through to 2000 rpm. I have been trying to correct a rough idle issue on my 120r/ 660sm etc. I have auto tuned it several times with adaptive switched off and I have attempted to set the target afr higher and higher but I still get a richened fuel table at idle. I attempted to drop some points out of my warm up enrichment but I still see this anomaly with the rough idle.  My map is 35- 40 Kpa at idle( 1016 rpm. ) any thoughts?

Idle is a rough spot for alot of auto tuners.  Try lowering your CDE by 10% and see how it does.  There is also a feature of setting min map closed loop works.  Setting that to something like 39 might help.  There is also a closed loop learn rate, TT is set to 2 from the normal 1.5 but setting it to something like 1.8 i have found to help slow things down at idle so it's not so back and forth.

[joe_lyons]

What is your map load value at 1000 rpm?  42 is ok also but 35ish sounds more like it. 

You try to keep the map load reading of 1 at idle. 

IVO and IVC will not affect WOT.  Changing from a value of 1 to .98  should lower the WOT map value by 2 kpa (100->98).

Lower your IVC number and see what your idle map does.

This is my current Map Load table, wich achieve a nearly straight MAP at WOT accross the rpms range (corrected for barometric/elevation). I think i'm right here...
10001,00
15001,00
20001,00
25001,00
30000,98
35000,96
40000,95
45000,96
50000,97
55000,98
60001,02
65001,05
70001,00
75001,00
80001,00
85001,00
90001,00

Do you think the IVC number is incorrect for this cam ? If the IVC/IVO are good for this cam and work for everyone but me, then the problem may be elsewhere, like an intake leak ?

Every motor is a bit different.

[Frédéric CM]

got my IVC set to 29 by Dynojet, will try it
Jamie went the other way and set the IVO to 7... what difference can i expect between both ? and how do you calculate IVO/IVC ?

Thanks !

[Sunny Jim]

So Joe, chiming here, I read my map load normalisation which I understand to be similar to bias global. At 1000 rpm my ratio is 1.02 through to 2000 rpm. I have been trying to correct a rough idle issue on my 120r/ 660sm etc. I have auto tuned it several times with adaptive switched off and I have attempted to set the target afr higher and higher but I still get a richened fuel table at idle. I attempted to drop some points out of my warm up enrichment but I still see this anomaly with the rough idle.  My map is 35- 40 Kpa at idle( 1016 rpm. ) any thoughts?

Idle is a rough spot for alot of auto tuners.  Try lowering your CDE by 10% and see how it does.  There is also a feature of setting min map closed loop works.  Setting that to something like 39 might help.  There is also a closed loop learn rate, TT is set to 2 from the normal 1.5 but setting it to something like 1.8 i have found to help slow things down at idle so it's not so back and forth.

So Joe, chiming here, I read my map load normalisation which I understand to be similar to bias global. At 1000 rpm my ratio is 1.02 through to 2000 rpm. I have been trying to correct a rough idle issue on my 120r/ 660sm etc. I have auto tuned it several times with adaptive switched off and I have attempted to set the target afr higher and higher but I still get a richened fuel table at idle. I attempted to drop some points out of my warm up enrichment but I still see this anomaly with the rough idle.  My map is 35- 40 Kpa at idle( 1016 rpm. ) any thoughts?

Idle is a rough spot for alot of auto tuners.  Try lowering your CDE by 10% and see how it does.  There is also a feature of setting min map closed loop works.  Setting that to something like 39 might help.  There is also a closed loop learn rate, TT is set to 2 from the normal 1.5 but setting it to something like 1.8 i have found to help slow things down at idle so it's not so back and forth.

[Gordon61]

CDE helps with valleys and peaks.  Also with abrupt percentage changes in VE from one cell to another vertically.

What did your VE numbers do when you lowered the CDE 30%?  I don't use it just to change MAP reading.

Thanks Joe, I've heard that one before and that was my original thinking.
When I put the 57H into my 2014 the VE tables dropped dramatically in the 0% TPS column up to about 2250 rpm and resulted in a steep climb from 0 to 7%.  This gave some steps between adjacent cells anywhere from 5 to 10 of a difference in that depression ...I thought this was what was making the bike jumpy at those low revs.

Without knowing what I was doing as such I used the CDE tables from the 2015 103 HO maps which have higher CDE values ...this raised the VE in that area and the VE tables looked better and the bike seemed to run smoother.  This was when I noticed that the MAP reading had changed from 40 to 30.

However, I'm now trying lower CDE values again, but the front and rear are more similar compared to 2014 stock...
The horrible dip at the 0/0 RPM/TPS corner of the VE tables is back (numbers down as low as 50 with the hills in the rest of the table maxing out about the 122 mark.  The MAP at idle is back up at 40.  BUT... it is running even smoother (I think) than I had with the CDE higher.

It's just all rather horribly confusing 

A bit of discussion elsewhere and I'm wondering if the PV is doing something different with CDE and MAP than the TTS for example.  In the PV it almost looks as if the CDE tables are another bias table for the MAP sensor, that you can change the values and see the MAP reading change??

Just to explain that: No matter what I set the CDE tables to high or low, to give a MAP at idle either 40 or 30 - the PV MAP Sensor Voltage is pretty much the same regardless.  It swings around a bit, as you might expect depending upon sample rate and when the reading is taken within the 4-stroke x 2 cylinders cycle.  But, what remains constant is the regularly lowest voltage on any of the log traces is always 1.6 volts ...according to a delphi MAP sensor spec sheet, I think 1.6 volts equates to somewhere in the high 30s ...see what I'm getting at??

[Frédéric CM]

how was the MAP with modified IVC/IVO and stock 2014 CDE ?

[Gordon61]

how was the MAP with modified IVC/IVO and stock 2014 CDE ?

If you mean my experience - the map I started with I got from someone else running a 57H and came from Dynojet.  If I remember rightly I think the MAP at idle was about 43 or so (I'm pretty close to sea level by the way). 

I just struggled to get it dialled in, autotune would take the numbers one way, then the next day autotune would take them the other way, and low speed running was really jumpy at anything under 2K.  That seemed awfully poor for a reasonably mild cam and I got to wondering if the MAP was being read from the sensor at the wrong time (which is where your Intake valve opening and closing settings come in).

Emailed Dynojet and explained and they sent me a new map. They had changed something because the MAP at idle came down to 38-40 or so and the bike dialled in much better and was reasonably smooth if you kept it above 16/1700 or so.  Note the CDE tables were still the same as stock for the 2014 I had.

I've been learning, testing, and tweaking CDE since then and it will now run reasonably well down to about 1400 or so in 3rd, anything less that that and it gets a bit too jumpy and you have to feather the clutch.  My idle is still a tiny bit jumpy compared to a couple of other bikes I was listening to at the dealers.

If I knew a bit more, I think I could get it better still ...didn't really want to have to buy a TTS or travel across the country to get a dyno tune tho, but it may come to that 

[Frédéric CM]

From PVtech @ Dynojet :
"My Harley expert looked at your situation and his comments were to have the IVC set at 30...  And, he said that changing the IVO from an 8 will only make things worse.  He also wanted me to assure you that a MAP reading of 42 is NOT all that high...  It's kind of borderline.  He said that if you live near sea level, it is not uncommon to see a MAP reading like that."

I did try with the IVC set to 29 instead of 30, idle only went down about 2kpa but idle quality was not as good as previously...
The MAP in your topic on the other forum have IVC/IVO set to 29/7. I can email you the map I have with 30/8, paste your VEs and CDEs/map load and give it a try

What does the CDE of HO cams look likes ?

[hdmanillac]

In every calibration file you have a hidden table called "map default table". I think it is used when the MAP sensor failed.

I suggest you to ask Dynojet to send you a copy of this table.

Then do some logs with a lot of decelerations and compare the MAP you obtain during decelerations with the map default table (columns 3, 4, 5% TP).

Adjust CDE to get map decel curve corresponding to map default table.

After that you will get a smoother bike, stable calibration and iddle and a good MPG (if your afr table is not to rich).

[Frédéric CM]

email sent

[Gordon61]

Never heard of that one, thanks for the new info hdmanillac

email sent

Let me know what you get pleeeeeassse 

You mentioned tooth numbers for IVO and IVC ...can you see/change these or is this just what Dynojet are telling you?

I just got a map and have no idea what my numbers ore set at ...I did wonder about asking to try one either side of what I have but never got around to it.

cheers

[Frédéric CM]

Nope I ask the PVtech to know the settings... you should try emailing him

[Gordon61]

Nope I ask the PVtech to know the settings... you should try emailing him

Knowing what the number was, wasn't any use if I couldn't see or change it so I never bothered to ask, but some of the pros do have the software ...was why I wondered if you had it. 
No doubt once DJ work out how to wrap it all up for us DIY types there will be an update to WinPV or something.

The CDE for the 103 HO 2016's (think I said 2015 earlier sorry) is
Front
140
140
140
121
104
92
79
73
67
61
51
43
35
31
31
31
31

Rear
115
115
115
105
97
90
81
72
61
48
36
33
31
31
31
31
31

[Frédéric CM]

If a pro is willing to share the super-version of the software, I'm here !!

This CDE is what i found on a 2016 softail (DJ tunedb), but is different than a 2016 touring... both sahre the HO cams ?

[Gordon61]

I've looked at so many I'm losing the will to live.  My guess would be significantly different exhaust designs between the two platforms, hence different reversion characteristics

[Sunny Jim]

In my case, I feel reversion is a player in this. Whilst my overlap is only around 38, I do have 660 lift. I figure the way these Vtwins work and the inlet manifold design, I probably have a fair bit ov movement from one cylinder to the other. Surely this would affect idle. No matter what I do - A/Ts, Ve adjustments, new seals on manifold etc , I still have slightly richer event on the rear cylinder and via my data logs- MLV , I have a rougher idle on the rear than the front.

[Frédéric CM]

Attached is the Map default table.
At 0% TP, my MAP is far from this
At 100% my MAP isn't far from this, before a flattened the "wave" with the Map Load Normalisation

[Gordon61]

In every calibration file you have a hidden table called "map default table". I think it is used when the MAP sensor failed.

I got to thinking about this.
If this is a default, backup table in the event of something going wrong ...i.e. a failed MAP sensor, I would have to presume that it was based on the original engine characteristics, no?

If you change exhausts and especially cams, you can dramatically alter the characteristics of the engine, event timing, reversion, flow dynamics etc  This is why we need to re-tune the ECU.

In simple terms it is how much air and how much fuel - surely the MAP sensor voltage (ok with a sensor bias table) is what should be used in the fuel calculation?

So, in the first place, why are we messing around with a calculated MAP value reading? What is the PV doing with that number?

Fiddling with a table (CDE) to make the MAP reading say something other than what the actual pressure is doesn't seem to make sense

And if there are hidden tables designed for limp mode get you home stuff, presumably in an ideal world THEY would need to be changed to reflect the new engine characteristics, rather than the other way around??

Just thinking aloud

[ben31]

MAP value does not reflect exactly engine characteristics. It is a physical/real measure and you need a sensor to get it.

You can guess an average value, using a table, but with the same bike, in two places with different elevation, at the same TP/RPM you won't have the same MAP.
Or even, in your place when atmospheric pressure changes...

[Gordon61]

Just for info, I was using the following CDE tables which on my bike with a 57H and Pro Pipe 2-1, and this gave a MAP at idle of about 31

Front, 153, 153, 150, 133, 114, 96, 79, 73, 67, 61, 51, 43, 35, 31, 31, 31, 31
Rear, 110, 110, 109, 105, 97, 90, 81, 72, 61, 48, 36, 33, 31, 31, 31, 31, 31

[Gordon61]

MAP value does not reflect exactly engine characteristics. It is a physical/real measure and you need a sensor to get it.

You can guess an average value, using a table, but with the same bike, in two places with different elevation, at the same TP/RPM you won't have the same MAP.
Or even, in your place when atmospheric pressure changes...

Exactly what I was getting at ...there IS a sensor so why is the PV calculating a different MAP value to what the sensor is detecting ...at different engine characteristics, and altitude and different weather on different days

[Frédéric CM]

Just for info, I was using the following CDE tables which on my bike with a 57H and Pro Pipe 2-1, and this gave a MAP at idle of about 31

Front, 153, 153, 150, 133, 114, 96, 79, 73, 67, 61, 51, 43, 35, 31, 31, 31, 31
Rear, 110, 110, 109, 105, 97, 90, 81, 72, 61, 48, 36, 33, 31, 31, 31, 31, 31

With your tune that has IVC/IVO modified by Dynojet (12D103_57H_ProPipeQuiiet) or not ?
I'll give a try to 2016 softail HO CDE table to see what happen to the idle kpa...

Just to remember, why are we targeting 35kpa at idle ?

[ben31]

...there IS a sensor so why is the PV calculating a different MAP value to what the sensor is detecting ...

Maybe I'm wrong, but here's how I think it works:

The MAP value is used to calculate how much air is flowing into the engine.
The MAP SENSOR reads the real MAP value

BUT the air volume and the MAP reading are biased due to a reversion effect.

The CDE table estimates this bias amount
Using the CDE you calculate the MAP without the reversion disturbance
With this calculated MAP, you can calculate the VE taking in account only useful fresh air

[Sunny Jim]

Joe, you mentioned some possible changes to CDE. You a
Do mentioned altering the parameters of closed loop regarding min closed loop works and learn rate.
Can you guide me through the PV and target tune as to where these alterations can be made please?

[joe_lyons]

MAP value does not reflect exactly engine characteristics. It is a physical/real measure and you need a sensor to get it.

You can guess an average value, using a table, but with the same bike, in two places with different elevation, at the same TP/RPM you won't have the same MAP.
Or even, in your place when atmospheric pressure changes...

Exactly what I was getting at ...there IS a sensor so why is the PV calculating a different MAP value to what the sensor is detecting ...at different engine characteristics, and altitude and different weather on different days

The ecm does this because with a very short intake track and an odd fire the MAP signal goes crazy up and down.  The MAP tooth read/poll close and open tell the ecm exactly when to look at the map sensor.  This is usually based off of the cam timing events.  The MAP load normalisation table is in essence a calibration bias table to make sure the MAP sensor is on the right track.