Turbine rpm sensor logging

[Wolf_Tm250]

Hi All,
need help to setup the turbine rpm sensor in my M400.

Data from manufacturer.
----------------------------------------------------------------------------------------------------------------
The sensor is a TS180.

Red +12V supply (min 8V max 15 volts) at approx 30mA
Black Ground (negative battery terminal)
White Data Logger / Tachometer drive (blade frequency divided by 10N where ‘N’ can be programmed between 1 and 32. Open collector output with 2K2 pullup to +12V

Use with Data Loggers
A turbine speed of 180,000 rpm with a 14 blade turbo has a blade frequency of 42,000 Hz. The module divides this by 10 and then by ‘N’ to give an output frequency that can be fed into data loggers. The software settings required in the Data Logger software for some values of ‘N’ are shown below.

TS-180 Output Frequency
(14 blades at 180,000 rpm) Data Logger Software Settings to log as Turbo RPM
‘N’ TS-180 Output Frequency Teeth Multiplier
1 4200 Hz 14 10
2 2100 Hz 7 10
14 300 Hz 1 10

N can be set to other values if needed to better match the frequency range.
----------------------------------------------------------------------------------------------------------------------

As far as I understand i have to connect white wire to a digital input, given that it's a 5V square wave.
What else to do to setup it?
Is it correct to:
Type: 2 RPM Measure
Calibration: 4200 with "N"=1 ?
Active edge: ?
Noise window: ?
Noise increment: ?
Noise limit: ?

Thanks in advance.

[Scott@FP]

The ones we use its real simple, just a couple of absolutes- can't go over 6KHz at the ECU digital input. Can't display RPM's above 20,000. Can't have a speed/RPM cal value above 19999, and cal value must be a whole number from 1-19999.

Workaround is to configure your speed input to display in RPM x 1000 (or RPM x 100).

For instance in your dig input function setup, use RPM, and for a sensor that has an internal div/8 function and a display RPM in whole thousands the cal would be;

(sensed blade count/divider circuit output factor)*1000

14 blades/8 = 1.75*1000=1750

enter '1750' in digital input function/calibration field.

12 blades/10 = 1.2*1000 = 1200

etc.

The units we use output one 100uS 5v pulse per 8 blades, are 6mm x .50 threads and 40mm length.
(edit to correct thread pitch)

I've never had to use any of the noise parameters, the ONLY problems we've ever seen is from sensor to blade gap being too CLOSE, it then sometimes picks up stray 'doubles'.

[stevieturbo]

Just spied this

http://www.oxfordrfsensors.com/products.html

How much was the sensor Wolf ? Does it need any form of external controller ? Or just a power supply ?

[Wolf_Tm250]

Just spied this

http://www.oxfordrfsensors.com/products.html

How much was the sensor Wolf ? Does it need any form of external controller ? Or just a power supply ?

Yes, it's what I have...
I don't remember... 250£ plus shipping?
It has an external controller...

[Scott@FP]

The sensor we're using costs $225.00 USD in kit form, 3 wires +5v, 0v and signal, uses a standard AMP SSC connector. Electronics are contained in the molded connector.

[Wolf_Tm250]

The ones we use its real simple, just a couple of absolutes- can't go over 6KHz at the ECU digital input. Can't display RPM's above 20,000. Can't have a speed/RPM cal value above 19999, and cal value must be a whole number from 1-19999.

Workaround is to configure your speed input to display in RPM x 1000 (or RPM x 100).

For instance in your dig input function setup, use RPM, and for a sensor that has an internal div/8 function and a display RPM in whole thousands the cal would be;

(sensed blade count/divider circuit output factor)*1000

14 blades/8 = 1.75*1000=1750

enter '1750' in digital input function/calibration field.

12 blades/10 = 1.2*1000 = 1200

etc.

The units we use output one 100uS 5v pulse per 8 blades, are 6mm x 1.00 threads and 40mm length.

I've never had to use any of the noise parameters, the ONLY problems we've ever seen is from sensor to blade gap being too CLOSE, it then sometimes picks up stray 'doubles'.

Perfect, thanks!