PDA

View Full Version : How to log the duty cycle of the valve in flow meter?


5mall5nail5
27-05-2010, 01:36 PM
Hi all -

Just for my datalogging purposes I'd like to log what the injector is doing in the HFS-6 kit. I have a spare input on my DTA ECU that I can input to. This will give me a better idea of what my duty cycle curve is looking like out of the Aquamist vs my fuel injection duty cycle.

How can I log this? Thanks!

Howerton Engineering
28-05-2010, 08:01 PM
In the back of the manual there are pinouts for each device. You can find the 0-5v conditioned flow sensor signal on the FCM and log this.

5mall5nail5
29-05-2010, 06:32 AM
In the back of the manual there are pinouts for each device. You can find the 0-5v conditioned flow sensor signal on the FCM and log this.

Thanks Jeff - you mentioned this but I was looking on the main controller, whoops.

Thanks a bunch!

Is it the yellow "Flow Sensor Output signal" or brown "PWM Valve Distribution Signal" wire?

Richard L
29-05-2010, 09:15 AM
I think the signal you are looking for is the conditioned output of the flow sensor. Unfortunately it is not shown at the back of the page.

This signal is in the green connector (user port). Pin #4. Forth conenction, counting from the gauge side. You can get RJ45 plug and make you own logging connection.

The sensor voltage is as following:

- standby ........... ~0.5V
- span ........... 0.6 to 4.5V

you will priobably only read a sopan of 0.6- 3.5V because the signal is conditioned by the "SC" to display ~5-6 bar on the gauge.

Alternatively, It is possible to read the "RAW" signal but it will be in "pulsed" form. You have to use a "frequency" input on your logger. Also requires soldering work.

Howerton Engineering
29-05-2010, 02:45 PM
Thanks Richard.

5mall5nail5
08-10-2010, 04:48 PM
I think the signal you are looking for is the conditioned output of the flow sensor. Unfortunately it is not shown at the back of the page.

This signal is in the green connector (user port). Pin #4. Forth conenction, counting from the gauge side. You can get RJ45 plug and make you own logging connection.

The sensor voltage is as following:

- standby ........... ~0.5V
- span ........... 0.6 to 4.5V

you will priobably only read a sopan of 0.6- 3.5V because the signal is conditioned by the "SC" to display ~5-6 bar on the gauge.

Alternatively, It is possible to read the "RAW" signal but it will be in "pulsed" form. You have to use a "frequency" input on your logger. Also requires soldering work.

Richard I just did this and had some weirdness -

I have a CAT5 cable with this pinout:

http://i40.tinypic.com/es5feb.jpg

I am using the BLUE wire (positing 4 in the connector). When I hooked that up to an Analog input on my ECU, the gauge had full bars lit on the gauge with the engine off but ECU/etc on. I unplugged the cable from the USER port and the bars went off.

Is this not correct?

Howerton Engineering
08-10-2010, 05:55 PM
I'm not certain but I think when Richard says pin 4 counting from the gauge side of the controller that means pin 5 on your diagram the way the connector plugs into the controller.

5mall5nail5
08-10-2010, 07:44 PM
Jeff - I can move it to that pin. The gauge port on the unit is to the left of the user port when looking at it head on, so I counted over 4 pins. They coincide with the drawing above - if thats incorrect I can move it I hope no damage occurred.

Howerton Engineering
08-10-2010, 07:51 PM
Sometimes it's confusing. If you take the connector int he picture above and flip it over it plugs in going downwards. In this case counting from left to right I think pin 5 in your drawing is right. Maybe Richard can confirm.

5mall5nail5
08-10-2010, 08:03 PM
Hrm - should i try it? I don't want to damage anything. The gauge port is oriented opposite of the rest of the ports, so then maybe it is pin 5 if thats what Richard meant.

Howerton Engineering
08-10-2010, 08:08 PM
Fourth pin counting from the gauge side, if you turn the RJ connector over like you are plugging it into the controller that would be pin 5 on your drawing.

If you physically look into the RJ port on the controller and count 4 from the gauge side that's the one.

5mall5nail5
08-10-2010, 08:14 PM
Ok I still don't really understand "gauge side" but if its pin 5, I can deal with that. I am going to wire the white/blue wire as the flow meter conditioned signal.

Richard L
08-10-2010, 10:28 PM
Here is the correct signal for the flow sensor:

Pin #5 (blue/white) = calibrated flow sensor signal .........~0.5V to ~4.5V

5mall5nail5
09-10-2010, 02:13 AM
Got it working Richard thanks! Although its somewhat of an arbitrary value to log, I can see that the W/I is turning on and going.

Howerton Engineering
09-10-2010, 04:57 AM
You can put the system in test for one minute and measure the fluid sprayed while logging the voltage. This will give you a correlation of voltage to flow. Depending on how ambitious you are, you can map a few points and then have some data of what is going on.

Just remember, if you change the SC position on the gauge you will need to re-calibrate.

Richard L
09-10-2010, 09:51 AM
Got it working Richard thanks! Although its somewhat of an arbitrary value to log, I can see that the W/I is turning on and going.

The main aim of flow montoring is to detect flow anormalies rather to inform absolute flow. The SC allows users to normalise the flow across the 8 bars of display.

It works like a temperature gauge or tank level gauge on a dash board. The actual cc/min is not the primary objective. That said, Jeff suggested calibrating the sensor to display absolute flow if you want to construct your own failsafe strategy on your ECU.

It is posssible to extract the raw signal from the sensor pulse/sec and make your own "frequancy to voltage" decoder for your logger. This way you can have both.