richyb
|
 posted on 17/6/09 at 10:40 AM |
|
|
Megasquirt - bogging under hard throttle
Hi guys, one for the megasuirt experts here -
just fitted GSXR itb's to my 1800 zetec megasuirt set up. tuned in VE MAP etc using megalog viewer and LC1, and runs generally well under
moderate loads. problem starts when i open up the throttle quickly under load in 2nd gear and above - it boggs down, but will recover if i throttle
off and feed the throttle on more gently. Have tried varied adjustments to the accel enrichment to no avail.
my next theory is the fuel delivery - i have a cheap regulator with gauge (others have bought these on here off e bay) that i have set at 3 bar - is
this sufficient to deliver sufficient fuel under load / do i need to increase ? - didn't want to change this unless i have to as will have to
re-map etc.
no air filters fitted as yet - would this effect things that much ?
any advice / ideas apreciated.
|
|
|
|
|
omega0684
|
| posted on 17/6/09 at 10:51 AM |
|
|
i have the same problem with my set up at the moment,
firstly, i would fit an air filter (you don't want to be sucking in bees and suck like!)
secondly, your ITB's should be ok if you run them about 44psi, so adjust regulator accordingly.
thirdly, what is your AFR doing when you plant your foot? is it leaning off or richening up? according to the MS manual the 'bogg down effect
can occur if the mixture is too rich & too weak' it also mentions that the aim is to get a consistant AFR across the table and then use the
acc enrichments to alter the fueling where neccessary.
keep fidling with the acc enrichments if you are happy with your ve table. they are quite delicate so only increase or decrease in small amounts
|
|
|
BenTyreman
|
| posted on 17/6/09 at 01:08 PM |
|
|
You say it bogs down, but what does your datalog show? Does it go lean or rich?
Too rich tends to feel like the engine isn't responding to the throttle as fast as it should be. Too lean generally results in misfires.
Acceleration events usually last for around 0.2 of a second. Hesitation longer than that generally indicates a problem with that area of the map.
You don't say if your using Alpha-N or Speed Density. If your using Alpha-N then by "feathering the throttle", you could just be
using a part of the map that is correctly fuelled and avoiding part of the map at higher throttle opening that is incorrect.
Post your datalog (zip it up first to reduce the size) and MSQ and I can take a gander.
|
|
|
richyb
|
| posted on 17/6/09 at 01:38 PM |
|
|
air filter is a temp thing - hopefully one its way.
I am using speed density currently - idling well so have not switched to alpha N . datalog and mxq files attached - welcome any comments. feels like
engine is not responding quick enough / fuel shortage ? will definately increase the fuel pressure and try again with the accel settings.
thanks
|
|
|
richyb
|
| posted on 17/6/09 at 01:40 PM |
|
|
msq file attached also
|
|
|
BenTyreman
|
| posted on 17/6/09 at 02:16 PM |
|
|
First problem is that you seem to have a very noisy MAP signal. This can be fixed by putting a small fuel filter in the middle of the line.
Second problem is that you are using Speed Density. You have no control over the fuelling once the throttles are open more than about 10-20%. You can
see that as you open the throttle, the MAP immediately pegs to atmospheric. The mixture starts off correctly, but as you open the throttle further the
mixture leans out.
If you have the req fuel and opening time values correct, then the top of the VE table should be approximately 100. You have half this value. The
value of 13 for req fuel seems quite high. What size injectors are you running? With an idle pulsewidth of 3.4ms, I would switch over to 4
squirts/alternating or possibly even 4 squirts/simultaneous.
There also seems to be a lot of discontinuities in the data. For every 10 records (about 1 second of real-time), there is a 4 second gap. Do you use a
USB-serial adaptor?
There is a massive amount of noise on the batt V trace and a couple of resets. This could be causing the noisy MAP signal and the data drop-outs. Are
you sure you have a good clean power and earth feed to the Megasquirt?
Lastly, I can't see any acceleration events at all in the datalog you posted. The problem is that you have selected TPS-based acceleration, but
you have the TPSdot threshold at 31 V/s. I believe around 0.5 V/s is the default setting.
Hope that helps.
|
|
|
richyb
|
| posted on 17/6/09 at 03:19 PM |
|
|
thanks for this Ben - all extremely useful info
the injectors are GSXR 600 ones - best info on here is that they flow at 230.
yes i do use a USB serial adaptor - also has a USB extension lead in the line to the laptop - is this a bad option ?
I was fairly meticulous with the earthing etc when i installed , but will check this out.
accel TPSdot was turned up high for this log (so as not to effect VE tune), but have tried settings between 0.3 and 2.5 V/s since without any great
improvement.
thanks again for your help
|
|
|
BenTyreman
|
| posted on 17/6/09 at 03:38 PM |
|
|
230 sounds about right. Ideally you want to increase the number of squirts as it allows any changes in fuelling to happen quicker.
You need to track down the batt V problem. I have a rock steady voltage of around 13.4 V with a maximum swing of 0.2 V. With the voltage changing by
around 2 V on your datalog, and with a injector opening battery correction of 0.2 ms/V, you will find the injector pulsewidth varying by 0.4 ms when
it probably shouldn't be varying at all.
You don't need to disable the accel enrichments to do VE tuning. MegaLogViewer automatically removes data taken while accelerating.
One of the most important things to change is the control algorithm from Speed Density to Alpha-N. You will have to start from scratch on the VE
table, but the results will definitely be worth it. You can get a reasonable idle and cruise tune with SD on throttle bodies, but as soon as the
throttles open by more than around 10%, the MAP value no longer accurately measures the air flow into the engine.
Do you have any programs running on your laptop that might interfere with the USB-serial adaptor? The missing data isn't really a problem, but
it means that you aren't getting the full picture as to what is going on.
|
|
|
fatbaldbloke
|
| posted on 18/6/09 at 10:12 PM |
|
|
megasquirt
Absolutely agree with the comment above. You need to run AlphaN. I've put some notes on my website as well on setting up with GSXR TB's
that might help.(www.fatbaldbloke.co.uk).
|
|
|
paulf
|
| posted on 19/6/09 at 02:53 PM |
|
|
I used to run speed density and found that although it would run well on light throttle and cruising and also at full throttle it was impossible to
get a good throttle response when driving fast and on off the throttle.This made it difficult to balance on the throttle when cornering fast as it
tended to die then power come in all of a sudden. I have now swapped to Alpha N and now find it much better to drive but is now less economical at
cruising speeds although am still refining the map.
I also changed to RPM based throttle enrichments with Map triggering ,this seems to work better.
Paul
|
|
|
corpi
|
| posted on 20/6/09 at 09:56 AM |
|
|
I'm having the same issues, moving from closed to open throttle, gives problems.
Has anyone tried the MSII version 3.01? it is supposed to be able to do blended AlphaN and Speed density, and contrary to the olede versions, it can
do AlphaN at low speed, and speed density at higher (engine speeds)
It was made for engines that: "for when MAP peaks to 100 as soon as the throttle is cracked and there is no more load control unless you use
alpha-N. "
Seems to be made for this type of problem!
Anyone tried this?
Pieter
|
|
|
BenTyreman
|
| posted on 20/6/09 at 12:06 PM |
|
|
I'm running MS2Extra code. It has two tables, one for SD and one for Alpha-N, and they are multiplied together.
I found tuning it to be more trouble than it's worth. I've currently got a tune that I'm more than happy with, with just Alpha-N.
I might go back and revisit blended tuning to get a bit more fine control at cruise and hopefully a bit more economy, but I suspect the low fuel
economy is more due to the driving style than anything else.
|
|
|
chasmon
|
| posted on 21/6/09 at 10:20 PM |
|
|
I've been meaning to mess around with Hybrid Alpha-N for a while. This essentially adds a MAP term to the fuel calc.
The squirt is calculated by the normal method for Alpha-N but the final value is multiplied by a MAP/100 term. This should mean your WOT figures
remain unchanged but for small changes at light throttle the large change in airflow is "seen" by the MS.
My engine spits back at low RPM low throttle openings as the TPS simply doesn't have enough resolution at small openings as the TBs (GSXR1000)
have a shallow angle on the plenum meaning small change in angle gives huge changes in airflow
[Edited on 21/6/09 by chasmon]
|
|
|
BenTyreman
|
| posted on 21/6/09 at 10:29 PM |
|
|
quote:
Originally posted by chasmon
My engine spits back at low RPM low throttle openings as the TPS simply doesn't have enough resolution at small openings as the TBs (GSXR1000)
have a shallow angle on the plenum meaning small change in angle gives huge changes in airflow
That's exactly the reason I moved to MS2. I found there wasn't sufficient precision to measure the TPS sensor when pottering around car
parks and driving on and off the throttle through towns.
One of the problems I found with using a hybrid approach to fuelling was that the MAP lags the throttle opening when coming back on the throttle after
a gear change, resulting in lean misfires. Changing to pure Alpha-N sorted this out.
Incidently, the so called "hybrid Alpha-N" on MS1 is essentially the same as using the "multiple MAP" option on MS2. MS2 has
the ability for much more accurate control over the MAP/TPS interactions by allowing you to have tables for both MAP and TPS. The final pulsewidth is
calculated by either adding or multiplying the MAP and TPS tables.
|
|
|
fatbaldbloke
|
| posted on 22/6/09 at 06:13 PM |
|
|
quote:
My engine spits back at low RPM low throttle openings as the TPS simply doesn't have enough resolution at small openings as the TBs (GSXR1000)
have a shallow angle on the plenum meaning small change in angle gives huge changes in airflow
Absolutely correct, but there are easier ways to get around it...(I've written some more extensive notes on my website)...in the original GSXR
TB's there is a second butterfly, often wrongly referred to as the choke butterfly, which has to be removed when fitting to a car engine. On
the bike this serves to modulate the MAP change just off throttle, so it doesn't see the big MAP change you're experiencing. On a car
installation you're sort of stuck with the problem as there's no easy way of driving this second butterfly. However, the trick is to get
the active throttle butterfly to be a bit more open on idle so it's largely past the position where the rapid MAP change occurs. To do this
open the idle setting a bit and then retard the ignition to bring the speed back. You will probably have to richen the mixture as well, so it runs a
bit rich and retarded at idle. This will also do wonders for your throttle blip from idle. The price to pay is a bit less efficient at idle. To put
some numbers on it I would estimate running at idle at around 6-8 degrees BTDC on a Zetec (instead of the nominal and usual 10) and with a AFR around
12 - 12.5.
If anyone is in the Bedford area and wants proof that a standard MS1 V3 running extra code and pure AlphaN with bike TB's can be made to run
pretty much like a production system with no pops, bangs, etc provided you're prepared to put in the hours of logging, just drop me a mail. It
might lift the spirits in a low moment...
[Edited on 22/6/09 by fatbaldbloke]
[Edited on 22/6/09 by fatbaldbloke]
[Edited on 22/6/09 by fatbaldbloke]
|
|
|
chasmon
|
| posted on 22/6/09 at 06:42 PM |
|
|
Hi FBB, we exchanged emails a few times via the WSCC 
Thanks for the tip of retarding the ign and opening up the throttle. It is literally when feathering the throttle. Sometimes its fine sometimes the
car can jerk resulting in one of the those off the throttle on the throttle moments that makes you look a plumb outside of tescos.... 
PS any chance of your map for a gander?
charliewallis at hotmail com
[Edited on 22/6/09 by chasmon]
|
|
|
flak monkey
|
| posted on 22/6/09 at 07:23 PM |
|
|
Very interesting last post FBB, will be doing my duratec install soon and will bear that in mind. I have the same problem just off idle with my gsxr
600 throttle bodies on my pinto, still running dizzy on that though so dont have that same control over the idle advance unfortunately.
Sera
http://www.motosera.com
|
|
|
BenTyreman
|
| posted on 22/6/09 at 07:23 PM |
|
|
quote:
Originally posted by fatbaldbloke
To do this open the idle setting a bit and then retard the ignition to bring the speed back. You will probably have to richen the mixture as well, so
it runs a bit rich and retarded at idle. This will also do wonders for your throttle blip from idle. The price to pay is a bit less efficient at
idle. To put some numbers on it I would estimate running at idle at around 6-8 degrees BTDC on a Zetec (instead of the nominal and usual 10) and with
a AFR around 12 - 12.5.
This is a good trick in general, especially if you increase the advance to around 14-16 degrees on the row above idle. The switch from a retarded
advance at idle to the "correct" advance just off idle helps the engine to rev up faster. The larger idle throttle opening also decreases
the amount of engine braking and allows the engine to idle when cold without an idle valve.
I'm currently running 8 degrees at idle, with 20 degrees at 400 RPM and 16 degrees at 1% throttle. I currently have no problem idling at 14.7:1,
but I suspect it will vary from setup to setup.
|
|
|
fatbaldbloke
|
| posted on 22/6/09 at 09:44 PM |
|
|
quote:
Sometimes its fine sometimes the car can jerk resulting in one of the those off the throttle on the throttle moments that makes you look a plumb
outside of tescos....
Two things to help here. One, work at the map so there isn't a flat spot just off idle, as mentioned above. This will help hugely.
The other is try putting an extra throttle return spring somewhere near the pedal, so it doesn't add any load onto the actual throttle linkage
to the TB's but gives you something firmer to press against.
I'll send the current map by email shortly.
|
|
|
.jpg)
