Data Logger Graphs

Discussion on Data Loggers, how to read your logs, what to improve, etc.

Moderator: RPM

Post Reply
Message
Author
User avatar
John_Heard
Site Admin
Posts: 5734
Joined: Thu May 12, 2005 11:20 am
Location: Resume Speed, Kansas
Contact:

Data Logger Graphs

#1 Post by John_Heard » Mon Oct 24, 2005 9:53 am

I could have swore I had a post in here showing shift points. Randy and I were discussing this about a month ago how to determine when to shift but for some reason that post disappeared. Here's the graph we were discussing.

Image

This is an edited portion of the Accelerometer graph from my car on a motor only pass a few weeks ago. Randy you mind making a few comments on what you see here when you get a minute?

Guest

Accel shift points

#2 Post by Guest » Tue Oct 25, 2005 10:18 am

For those not familiar with the accelerometer, or "G Meter", it measures how hard the car is pulling. Since this is really the bottom line in drag racing, it is a pretty useful tool. To tune with an accelerometer, you simply do whatever you can to keep it as high as possible.

A typical acceleration graph should reach peak acceleration shortly after launch, usually within 2 to 3 tenths. (There are exceptions to this, one being small tire class cars that have more power than can be applied at launch.) After reaching peak acceleration, it will slowly drop off throughout the run, with "anomalies" at shift points.

What we see in this graph, I assume, is the middle portion of the run, with the 1-2 and 2-3 shifts. The two positive going spikes are the shift anomalies. If you look closely, just before the up spikes, you see a small down spike. This is present in most automatic trans shifts, and will show up on the graph unless the reading are filtered too much. When the shift occurs, the next gear clutches have to spin up the drum. This puts a load on the driveshaft, trying to slow it until the drum spins up. As this happens, the shift completes and the load starts pulling the engine down. At this point, the power being applied is the sum of the power being made by the engine, plus the inertia of rotating assembly. This produces the kick you feel, and shows up on the acceleration graph as a positive spike. It also typically produces a small amount of tire spin, which in this case is not a bad thing. This boost in acceleration will end when the engine reaches its minimum after the shift, and the graph readings will stabilize.

Determining the correct shift points is done by comparing the accelerometer readings before and after the shift, ignoring the anomaly created by the shift. For an optimum shift point, the line created by the accel readings before the shift should continue after the shift. In other words, if you lay a straight edge along the line before the shift, the line after the shift should also lay on the straight edge. If it is above the edge, you shifted too late, below the line, too early.

Following that line of reasoning, this graph shows a correct 1-2 shift, but the 2-3 shift is a bit early (we really need to see more of the graph after the 2-3 to be sure). In most applications, the car will want to shift the 2-3 at a higher engine RPM than the 1-2. This is because the engine is accelerating faster in first gear and more power is lost just accelerating the mass of the rotating assembly, so it wants out earlier.

The accelerometer does a very good job showing even minor shift errors, and the 2-3 shift error here will probably not cost even .003 seconds in ET. If you wait to shift till you feel it "lay over", you will see a major difference in the before and after shift alignment of the accel readings, and will also loose an unacceptable amount of ET.

Regards,

Randy

User avatar
John_Heard
Site Admin
Posts: 5734
Joined: Thu May 12, 2005 11:20 am
Location: Resume Speed, Kansas
Contact:

#3 Post by John_Heard » Tue Oct 25, 2005 10:40 am

Good info Randy, was wondering what caused the dip and spike.

Here's a more complete graph of that same run, I've edited it to fit on the screen better. Top section of the graph is Engine RPM and Driveshaft RPM, bottom is the acceleromter. Note: This is a motor only pass on a car with a nitrous gear and converter in it, hence the real soft launch.

Image

Stickshifter 55
Posts: 24
Joined: Wed Nov 21, 2007 4:16 pm
Location: NY

Re: Data Logger Graphs

#4 Post by Stickshifter 55 » Wed Oct 01, 2008 6:43 pm

Speaking of run graphs... This is a run from a week ago. The bumps in the DRPM are what lead me to buy a data logger in the first place. If it is not obvious from my screen name, this is a stick shift car. My plan is to soften the clutch in hopes of smoothing things out. I have some shock tuning to do as well. Observations welcome.
http://www.dailymotion.com/TMSjoe/video/x6v3pq_795_auto
Image

User avatar
CoMax Racing
Posts: 169
Joined: Sat Oct 01, 2005 4:40 pm
Location: Edmonton Alberta Canada

Re: Data Logger Graphs

#5 Post by CoMax Racing » Thu Oct 02, 2008 9:39 am

Regarding StickShifter55's graph.

Being new to data logging, I am also curious as to why the graph has two peaks at every shift point. Is that due to the time it takes to put the clutch in complete the shift and release the clutch again? Or is there another reason?

Corey

User avatar
John_Heard
Site Admin
Posts: 5734
Joined: Thu May 12, 2005 11:20 am
Location: Resume Speed, Kansas
Contact:

Re: Data Logger Graphs

#6 Post by John_Heard » Thu Oct 02, 2008 9:51 am

I don't think it's hooked up, looks like the tires are spinning. Like it's got the shocks set real tight or something.

Stickshifter 55
Posts: 24
Joined: Wed Nov 21, 2007 4:16 pm
Location: NY

Re: Data Logger Graphs

#7 Post by Stickshifter 55 » Thu Oct 02, 2008 10:45 am

My last race was in 2006 so I was a bit rusty. The double peeks are from over shifting. I want to shift @ 7500. The chip is set to 8500. The graph is telling me I'm against the chip early in the run. Later the car is spinning the tires.

Post race autopsy showed I had the shock extension set to max stiffness. That could be of the problem as John pointed out.

It is a clutchless transmission. Shifting is done simply by grabbing the next gear. No clutch is needed after launch.

Also my fuel system has an electric bypass that is controlled by the MSD. I set it to lean out @8000 thinking it would not get to that point and hence in my mind it was OFF. The data logger tells a different tail. This could also have some affect on the power.

Thanks for looking. I’m going to hit the track this Sunday if the weather is good and if I can get my NHRA license straightened out.

User avatar
CoMax Racing
Posts: 169
Joined: Sat Oct 01, 2005 4:40 pm
Location: Edmonton Alberta Canada

Re: Data Logger Graphs

#8 Post by CoMax Racing » Thu Oct 02, 2008 11:36 am

Is it a Lenco planetary, or a gearbox like a Jerico?

Corey

Stickshifter 55
Posts: 24
Joined: Wed Nov 21, 2007 4:16 pm
Location: NY

Re: Data Logger Graphs

#9 Post by Stickshifter 55 » Thu Oct 02, 2008 12:17 pm

CoMax Racing wrote:Is it a Lenco planetary, or a gearbox like a Jerico?

Corey
It’s a G-force 5speed. Just about the same thing as a Jerico or Liberty. A Lenco planetary transmission has an internal clutch in each stage. Therefore it is not considered clutchless. However you still shift it without using the clutch.

On my first post I have a link to an in-car video of the run represented by my data logger graph. You can see it in action.

User avatar
CoMax Racing
Posts: 169
Joined: Sat Oct 01, 2005 4:40 pm
Location: Edmonton Alberta Canada

Re: Data Logger Graphs

#10 Post by CoMax Racing » Thu Oct 02, 2008 12:35 pm

Yeah, the reason I was asking is I have been running a Lenco for years but never have had a logger on the car. I am putting one on next season and would like to gain as much knowledge as possible to help reading the graphs.


Corey

Post Reply

Who is online

Users browsing this forum: No registered users and 13 guests