Graphing RPM

Ask any questions about using PicoScope Automotive software here.

Graphing RPM

Postby mikehaimes » Mon Oct 06, 2014 7:18 am

I am wanting to graph engine rpm from a crank sensor either hall effect or inductive. The reason for this is to develop a test to ID which cylinder is misfiring. I have tried a math channel using freq(A)number of teeth*60 and this is what I got with a pressure transducer in cylinder 1, but it still does show cylinders 1 not contributing
Attachments
Nissan_Micra_2002_Petrol_20141005-0001.psdata
(1.16 MiB) Downloaded 286 times
mikehaimes
Newbie
 
Posts: 3
Joined: Sat May 10, 2014 4:05 am

Re: Graphing RPM

Postby Steve Smith » Mon Oct 06, 2014 11:24 am

Hello Mike and thank you for the enquiry.

I have used a maths channel 60/32*freq(A-2.5) for the digital crank signal on channel A.

I have then applied this maths channel and increased the scaling by a factor of 3 to reveal the drop in RPM for the cylinder containing the WPS (Cylinder 1)

Does this help with identification?

Take care......Steve
Attachments
Nissan_Micra_2002_Petrol_20141006-0001-REVISED MATHS.psdata
(1.17 MiB) Downloaded 259 times
Steve Smith
Pico Staff Member
Pico Staff Member
 
Posts: 426
Joined: Sun Aug 25, 2013 7:22 am

Re: Graphing RPM

Postby Fat Freddy » Mon Oct 06, 2014 11:46 am

Hi Steve

Just read a similar reply on iatn. When doing these calculations you have to allow for missing teeth. So on a 58 tooth crank it will actually be 60-2. So using your formula it becomes 60/60*freq(A-2.5) = 1*freq(A-2.5) = freq(A-2.5).

Make sense?

This should be 60/34*freq(A-2.5) (if I have counted correctly).

HTH
FF
User avatar
Fat Freddy
Zen Master
Zen Master
 
Posts: 267
Joined: Fri May 04, 2012 4:52 am

Re: Graphing RPM

Postby Steve Smith » Tue Oct 07, 2014 11:26 am

Hello FF and thank you for pointing this out. You have most certainly counted correctly as I have not allowed for the missing teeth.

The frequency counter is looking at intervals and whilst there are 32 intervals, there are also 2 x missing teeth that need to be considered.

I have used FF's maths channel 60/34*freq(A-2.5) as well as the one above to demonstrate the difference just 2 x teeth can make.

Thank you again, take care......Steve
Attachments
Nissan_Micra_2002_Petrol_20141007-0001-2 x maths channels.psdata
(1.16 MiB) Downloaded 241 times
Steve Smith
Pico Staff Member
Pico Staff Member
 
Posts: 426
Joined: Sun Aug 25, 2013 7:22 am

Re: Graphing RPM

Postby volrem » Tue Oct 07, 2014 6:45 pm

May be stupid but why freq(A-2.5) is better than freq(A)?
volrem
Guru
Guru
 
Posts: 146
Joined: Fri Jan 03, 2014 3:43 pm
Location: Estonia

Re: Graphing RPM

Postby Robski » Tue Oct 07, 2014 6:57 pm

volrem wrote:May be stupid but why freq(A-2.5) is better than freq(A)?


Sampling at the 2.5V threshold.
User avatar
Robski
Advanced User
Advanced User
 
Posts: 602
Joined: Tue Sep 14, 2010 6:36 pm
Location: Yorkshire

Re: Graphing RPM

Postby Fat Freddy » Wed Oct 08, 2014 7:52 am

Hi Steve

You have most certainly counted correctly as I have not allowed for the missing teeth.


I think I didn't. :oops: . My apologies. 2X 2 missing teeth. So 36 teeth. Which is a nicer number 10* each

Cheers
FF
User avatar
Fat Freddy
Zen Master
Zen Master
 
Posts: 267
Joined: Fri May 04, 2012 4:52 am

Re: Graphing RPM

Postby Steve Smith » Wed Oct 08, 2014 9:04 am

Thanks for the feedback on this one and there are no stupid questions Volrem, thank you for the post.

I have attached an application note regarding maths channels. Please take a read and feedback when you can.

Take care......Steve
Attachments
MATHS CHANNEL ARTICLE.docx
(2.42 MiB) Downloaded 297 times
Steve Smith
Pico Staff Member
Pico Staff Member
 
Posts: 426
Joined: Sun Aug 25, 2013 7:22 am

Re: Graphing RPM

Postby Valhalla » Wed Oct 08, 2014 9:14 pm

Fat Freddy wrote:Hi Steve

You have most certainly counted correctly as I have not allowed for the missing teeth.


I think I didn't. :oops: . My apologies. 2X 2 missing teeth. So 36 teeth. Which is a nicer number 10* each

Cheers
FF

I'd reitterate what FF says here - you need to use 36 as the number of teeth per engine revolution for this example, not 34 or 32. Fundamental to what FF is saying is that when you make a maths channel calculation for speed from a tooth frequency, you must use the tooth count as if there are no missing teeth, e.g. in this case add the 4 missing teeth to the counted 32 for a revolution. Then just reject the data at the missing teeth (drop-outs) in the speed calculation.

Am I missing something in the start of the thread here - all four cylinders DO look as if they are contributing equally, not the statement that No.1 is NOT contributing equally?
Valhalla
Advanced User
Advanced User
 
Posts: 66
Joined: Sat May 18, 2013 10:22 pm

Re: Graphing RPM

Postby volrem » Thu Oct 09, 2014 6:48 pm

Thanks Steve. I did read this article again and got some more info regarding math channels and understand some things better now.

I also did some tests and found no difference between these two options. However i believe crossing point in middle may give better results in some situations.

Also I must say that these new math channel functions are just great and make things much simpler.

Most of these are still not undestandable for me so I hope there will be some "real life" examples shown in the future. Or maybe I just have not noticed?
volrem
Guru
Guru
 
Posts: 146
Joined: Fri Jan 03, 2014 3:43 pm
Location: Estonia

Next

Return to PicoScope 6 Automotive software

Who is online

Users browsing this forum: No registered users and 0 guests