Homemade Pressure Transducers--Unexpected Results

More
1 year 11 months ago - 1 year 11 months ago #55812 by Joshasta
Recently made some pressure transducers and I am getting unexpected results. I am using USB cord for 5 volts DC. Confirmed with multi-meter. I have tried 2 different sensors. 1 is GM MAP sensor. The other sensor I tried is a Air Conditioning Pressure Switch. Both are giving me AC sine wave on the signal wire. I tried applying 10inHg to the MAP sensor, no change to the signal wire. What could be wrong here? I was reading this should work. I have confirmed wiring on All Data... What scale/probe do you guys usually select on the Verus when using these?


Last edit: 1 year 11 months ago by Joshasta.

Please Log in or Create an account to join the conversation.

More
1 year 11 months ago - 1 year 11 months ago #55813 by juergen.scholl
The frequency of the ac-signal seems to be 60 Hz. If so it probably stems from the USB charger, especially if it's present without applying pressure to the sensor. Try another voltage source, dc volts. The versus itself provides clean 5V from one of its pins on the extra plug (this also powers up the SO sensors). If needed I could search a related document. Or use a 7805 voltage regulator connected to the car battery.

Concerning the amplitude of the ac wave and the missing signal: you may have created some weird back feed/ground loop issue with this USB charger.... Where is your signal ground placed at and did the Versus run on battery or ac power supply?

An expert is someone who knows each time more on each time less, until he finally knows absolutely everything about absolutely nothing.
Last edit: 1 year 11 months ago by juergen.scholl.

Please Log in or Create an account to join the conversation.

More
1 year 11 months ago #55815 by Matt T
On the MAP sensor pic you don't appear to have a connection from sensor ground to scope ground so your signal is floating. Might not be your only problem but it needs fixed first. If you're still having problems after that move on to the other things Juergen mentioned.

Also FWIW I scoped a few wall warts before I found a good one with clean output.

Please Log in or Create an account to join the conversation.

  • Noah
  • Noah's Avatar
  • Offline
  • Moderator
  • Moderator
  • Give code definitions with numbers!
More
1 year 11 months ago #55816 by Noah
I had problems powering mine via a "wall wart" as Matt T calls them.

"Ground cannot be checked with a 10mm socket"

Please Log in or Create an account to join the conversation.

More
1 year 11 months ago #55819 by Joshasta
Thanks a lot for all the help everyone, I really appreciate it! I was using a 12v battery for scope ground. I moved my scope ground to sensor ground, and I got a clean signal of 4.5v atmosphere pressure. Vacuum pulls it down. Here it is with 15inHg reading a clean 2 volts.
I will be testing the AC Pressure Switch with air pressure today and hopefully doing some in-cylinder pressure tests. I will be using the same switch out a Nissan Altima as you have mentioned Noah.

Please Log in or Create an account to join the conversation.

More
1 year 11 months ago #55829 by Joshasta
I am using AC Pressure Switch from a 2008 Nissan Altima with the 4 cylinder. AllData shows the wiring schematic being the exact same for 4 cylinder and V6. I was hoping to get similar results as Noah did using switch from the 2010. All of these are from the same Nissan 4th Gen Altima.
I got nice linear rise in voltage from 50-100psi testing it with my leak down tester. But I didn't get a good waveform in-cylinder. I tried different settings. This is the best I got. Any other suggestions for probe selection or anything?

Please Log in or Create an account to join the conversation.

More
1 year 11 months ago #55830 by Matt T
If you're using a compression tester hose it might have a check valve or schrader in it.
The following user(s) said Thank You: Noah

Please Log in or Create an account to join the conversation.

More
1 year 11 months ago #55835 by Joshasta
That was it... Schrader valve in compression hose. I was able to get this waveform after removing it. Pretty good, but still some noise in the signal. I wasn't able to power my usb cable using the Verus. Not sure why. Maybe my USB ports need checking on it. So this is taken with usb cable plugged into wall. I wonder how this would work using 9 volt battery?

Please Log in or Create an account to join the conversation.

  • Noah
  • Noah's Avatar
  • Offline
  • Moderator
  • Moderator
  • Give code definitions with numbers!
More
1 year 11 months ago #55840 by Noah
Cool, glad you got it to work!
That's an interesting waveform, is there any running issue in that engine?

"Ground cannot be checked with a 10mm socket"

Please Log in or Create an account to join the conversation.

More
1 year 11 months ago #55871 by Paul P.
Well Noah, here is my analysis.

Granted I couldn't perfectly match the cylinder chart of due to the angle of the picture!


Never stop Learning.
The following user(s) said Thank You: Noah

Please Log in or Create an account to join the conversation.

More
1 year 11 months ago - 1 year 11 months ago #55872 by Joshasta
That waveform is during running compression. I just wasn't catching good waveforms during a crank. I would get a pressure spike but no other details.
I have been studying in-cylinder waveforms, I have the Microsoft overlay program, and I have noticed that area circled, exhaust valve closing to slowly it looks like? And also I have been looking at the pockets. The expansion pocket doesn't appear to be as deep as the intake. That may be pointing towards timing problem I am reading?
The engine doesn't have any major running problems. Fuel trims and power is good. no engine codes. It has an exhaust leak on a different cylinder... Engine does have 200K. I am going to check a different cylinder soon
Last edit: 1 year 11 months ago by Joshasta.

Please Log in or Create an account to join the conversation.

More
1 year 11 months ago #55873 by Joshasta

Please Log in or Create an account to join the conversation.

More
1 year 11 months ago #55874 by juergen.scholl
There is an online class taking place with John Thornton on May 3rd and 5th at Automotive Seminars. Total time ~ 4hours, cost 120 USD.
It is called Engine Mechanical Testing 2, covers in-cylinder text Ng as well as intake, exhaust and crankcase pressure testing. He uses case studies to explain theory and practice of the diagnostic proces.

I do not hesitate a second to recommend this class to any one interested in this topic.

automotiveseminars.com/online-training-events/

An expert is someone who knows each time more on each time less, until he finally knows absolutely everything about absolutely nothing.

Please Log in or Create an account to join the conversation.

More
1 year 11 months ago #55875 by juergen.scholl


Was this capture made with an 8 bit vertical resolution or is it just the response time of your sensor?

Anyhow, you may want to crank up the sampling rate significantly.

An expert is someone who knows each time more on each time less, until he finally knows absolutely everything about absolutely nothing.

Please Log in or Create an account to join the conversation.

More
1 year 11 months ago #55877 by Paul P.

and I have noticed that area circled, exhaust valve closing too slowly it looks like?


Exhaust Valve Close is not a good marker to use, because it happens during the overlap with the Intake valve. This is an Intake valve opening up slowly. This really shouts out a lifter issue, as opposed to a spring issue.

The Exhaust Valve Open and the Intake Valve Close are usually good markers to use to determine timing issues. Yours look reasonable.

Your Towers are leaning, and your pockets are uneven. This signifies a volume loss. This cylinder will have lower compression than the other.

Perform a Relative Compression test with an Ignition sync, I think this cylinder would be lower on compression.

What would put the nail in the coffin would be to have an FLS ( Piezo Pulse Sensor ) in the Intake with an ignition sync on this one!

Here is some great info on In-cylinder Diagnostics with none other than Brandon Steckler himself:



And the legendary mad scientist Bernie Thompson:


Never stop Learning.

Please Log in or Create an account to join the conversation.

More
1 year 11 months ago - 1 year 11 months ago #55889 by Joshasta
Really cool advice guys! Yes, I have been watching bernie, brandon, jerry, and as much of the tst seminars as I can. I will def look into the paid seminar coming up. I'm going to have to read over the comments in order to absorb it all. I just wanted to post a couple pics. This is just so cool picking up these signals with scrap sensors. I added intake pulls on channel 2 and ignition sync on channel 3. Added the layover. Pic of bank 1 and Bank 2. I see some differences. I have a lot to research. And yes, the pico pics are from the cheap little 2204A. I am just learning to use the software. The pico auto scopes are backordered. semi-conductor shortage... I cant afford the pico kit, I dont need any leads...

Last edit: 1 year 11 months ago by Joshasta.

Please Log in or Create an account to join the conversation.

More
1 year 11 months ago #55896 by Paul P.
@Joshasta

Kudos to you for doing for building and experimenting. Airflow diagnostics is an Awesome tool for non-intrusive engine mechanical testing.

Many of us have been down this road, and it was a great learning experience.

I noticed this flat spot on the waveform:


Here is a good fix for a homemade sensor, all it does is allow some of the signal voltage to be 'interpreted' as a vacuum. In other words, if your sensor outputs 0.5-4.5v, it would set the 'at rest/zero pressure' at ~1.5v. (You would need to measure the voltage you get if you intended to scale it)



Denis solves this flatline issue, and he's got some other interesting little builds in his gallery.

Always keep in the back of your head with pressure transducers and in-cylinder analysis;

1. Pressure Increases>>>Volume Decreases
Pressure Decreases>>>Volume Increases

2. Rate of Pressure Change.

-A good example is a leaning tower, assuming the piston speed was very constant, in your photo, pressure built faster on the compression stroke, and on the power stroke pressure is decreased slower.
Why is this? Volume is key. The cylinder inhaled less air than it normally would have so the pressure increased rapidly. Because of this smaller volume, the pressure decreases slower on the power stroke and an exhaust pocket does not get formed because the Exhaust valve opened when it should.

-when the intake valve opens we should see a sharp, almost straight down drop in pressure going in to the Intake Pocket. In your photo the rate of change is slow to decrease the pressure. This strongly suggests less volume can be inhaled by that cylinder. Remember the pressure is decreasing slower, therefore the volume must be increasing slower as well!

An FLS sensor, another super easy build, would show this cyclinder having a weak intake pull when that valve opens.

Most of all, have fun, take lots of known good captures, then the bad one's will stick out like a sore thumb!!

Never stop Learning.

Please Log in or Create an account to join the conversation.

More
1 year 10 months ago #56011 by Joshasta
@weycraze

Thanks for that advise, will give it a look over forsure!
The following user(s) said Thank You: Paul P.

Please Log in or Create an account to join the conversation.

More
1 year 10 months ago - 1 year 10 months ago #56316 by Joshasta
I've made this pressure sensor using a piezo element. I've Included pics of running compression, spark trigger sync, and intake and exhaust. Any tips on how I can better have this adjusted or set up? I am in the process of looking through all this data. What do you guys see from the pics? First pic is intake, second pic is exhaust. Looks like #1 isnt sucking in as much air.



Last edit: 1 year 10 months ago by Joshasta.

Please Log in or Create an account to join the conversation.

More
1 year 10 months ago #56326 by Paul P.
Johasta, Nice Captures you took. To obtain useful diagnostic valve movement it is best to use the piezo (FLS or Pulse) sensor while the engine is cranking.

Also, it's good to isolate the crankcase from the intake manifold, so block off all PCV components.

Keep the throttle plate closed when testing to get defined intake valve movements.

That's cool to see the in-cylinder and the EVO event line up together!!

Hopefully you can read these pics,




It's good to note that running exhaust FLS waveform can also identify a "miss", because no combustion means a vacuum across the face of the cylinder, just as a cranking waveform wold be!

Nice work.

Never stop Learning.

Please Log in or Create an account to join the conversation.

Time to create page: 0.305 seconds