382 replies to this topic

[ultimate]

Yes, makes sense to me. Which brand of analog oiltemp and oilpressure sensor do I need? What do you think how long do you need to bring that expansion-box?

[garw]

Expansion box is a quick turnaround, the release will wait for the testing of LCD cluster testing to be done . So after release of the cluster the box will follow about 4 weeks later . I don't have a particular sensor to recommend, I would assume the sensors sold by Elise websites Have been proven , ensure the oil pressure sensor is fitted on a remote oil line to reduce vibrations They are mechanical devices and vibrations will cause erratic readings and significantly Reduce its life I won't provide links to sensors in order remain neutral on products which are not my own . I think that's the forum rules for people like me .

[ultimate]

I am just asking because I think different brands will have different values/resistance-curves. For example: sensor a has 1000ohm at 100 degrees and Sensor b has 800ohm at 100 degrees

[Chris P Duck]

I am just asking because I think different brands will have different values/resistance-curves. For example: sensor a has 1000ohm at 100 degrees and Sensor b has 800ohm at 100 degrees

As long as the output is reasonably linear (over the operating range) I'd assume this doesn't matter and you'll be able to set it up as y = f(x) + c Amazing work. Keen to hear more details on the logging when you get a chance. I.e. Are ALL inputs logged? (Standard such as rpm and the 16 analogue inputs if the extra box is purchased?) Are calculated values logged? (G-force etc) What frequency are they logged at? >10Hz would be cool

Edited by Chris P Duck, 04 April 2017 - 07:46 PM.

[garw]

ok

 

gauge type oil pressure sensors which have a 3-160 ohm range     for 0-10bar   are good

 

oiltemp sensor should be gauge type

 

gauge type sensors have a lower resistance to ecu type

 

the 1000 and 800 ohms at 100 deg are too high 

 

100 deg should be under 100ohms

 

VDO gauge type for example would be around 50 ohms at 100 deg

 

 

 

 

[garw]

 

I am just asking because I think different brands will have different values/resistance-curves. For example: sensor a has 1000ohm at 100 degrees and Sensor b has 800ohm at 100 degrees

As long as the output is reasonably linear (over the operating range) I'd assume this doesn't matter and you'll be able to set it up as y = f(x) + c Amazing work. Keen to hear more details on the logging when you get a chance. I.e. Are ALL inputs logged? (Standard such as rpm and the 16 analogue inputs if the extra box is purchased?) Are calculated values logged? (G-force etc) What frequency are they logged at? >10Hz would be cool

 

 

The current datalogging implementation logs everything including gps coordinates and 3 axis gsensor (built in) its set to 10hz (just thought its was a good place to start)

 

the data channels are logged not the sensors themselves ,  ie throttle position value is logged 0-100%   not the analogue input its connected to ,  as it could be from the canbus.

 

gforce is logged in G's 

 

 

[Chris P Duck]

 

I am just asking because I think different brands will have different values/resistance-curves. For example: sensor a has 1000ohm at 100 degrees and Sensor b has 800ohm at 100 degrees

As long as the output is reasonably linear (over the operating range) I'd assume this doesn't matter and you'll be able to set it up as y = f(x) + c Amazing work. Keen to hear more details on the logging when you get a chance. I.e. Are ALL inputs logged? (Standard such as rpm and the 16 analogue inputs if the extra box is purchased?) Are calculated values logged? (G-force etc) What frequency are they logged at? >10Hz would be cool  

  The current datalogging implementation logs everything including gps coordinates and 3 axis gsensor (built in) its set to 10hz (just thought its was a good place to start)   the data channels are logged not the sensors themselves ,  ie throttle position value is logged 0-100%   not the analogue input its connected to ,  as it could be from the canbus.   gforce is logged in G's     

Thanks, I'd assumed G-force was calculated from the GPS rather than measured from a sensor. This really is good value when you compare it to a race logic performance box and micro-input module (which is what I currently have) Just a thought on expanding your market. Is all the hardware already there for recording track footage with a data overlay?

[ultimate]

ok   gauge type oil pressure sensors which have a 3-160 ohm range     for 0-10bar   are good   oiltemp sensor should be gauge type   gauge type sensors have a lower resistance to ecu type   the 1000 and 800 ohms at 100 deg are too high    100 deg should be under 100ohms   VDO gauge type for example would be around 50 ohms at 100 deg     

Will the "translation" (resistance = shown value) be programmable by the user himself? I had three different gauge oil temp sensors in the last years. All had different resistance at the same temp. Cheers

[garw]

ok   gauge type oil pressure sensors which have a 3-160 ohm range     for 0-10bar   are good   oiltemp sensor should be gauge type   gauge type sensors have a lower resistance to ecu type   the 1000 and 800 ohms at 100 deg are too high    100 deg should be under 100ohms   VDO gauge type for example would be around 50 ohms at 100 deg     

Will the "translation" (resistance = shown value) be programmable by the user himself? I had three different gauge oil temp sensors in the last years. All had different resistance at the same temp. Cheers

Yes

[garw]

 

I am just asking because I think different brands will have different values/resistance-curves. For example: sensor a has 1000ohm at 100 degrees and Sensor b has 800ohm at 100 degrees

As long as the output is reasonably linear (over the operating range) I'd assume this doesn't matter and you'll be able to set it up as y = f(x) + c Amazing work. Keen to hear more details on the logging when you get a chance. I.e. Are ALL inputs logged? (Standard such as rpm and the 16 analogue inputs if the extra box is purchased?) Are calculated values logged? (G-force etc) What frequency are they logged at? >10Hz would be cool  

  The current datalogging implementation logs everything including gps coordinates and 3 axis gsensor (built in) its set to 10hz (just thought its was a good place to start)   the data channels are logged not the sensors themselves ,  ie throttle position value is logged 0-100%   not the analogue input its connected to ,  as it could be from the canbus.   gforce is logged in G's     

Thanks, I'd assumed G-force was calculated from the GPS rather than measured from a sensor. This really is good value when you compare it to a race logic performance box and micro-input module (which is what I currently have) Just a thought on expanding your market. Is all the hardware already there for recording track footage with a data overlay?

Maybe , when the video input is sorted then it could be looked at .

[ultimate]

Okay, so for the records ;-) : I found the following values for VDO Sensors: Oiltemp for 150 degree Sensor: 60°C - 221,2 Ohm 90°C - 83,0 Ohm 120°C - 36,5 Ohm Oilpressure for 0-10bar Sensor: 0 bar - 10 Ohm 2 bar - 52 Ohm 6 bar - 124 Ohm 10 bar - 184 Ohm

Edited by ultimate, 04 April 2017 - 08:31 PM.

[ultimate]

found more values for the oiltemp-sensor.

 

Like every NTC it is not linear, see attached image FYI ;-) Is this non-linearity a problem for the cluster?

Edited by ultimate, 05 April 2017 - 06:45 AM.

[ultimate]

Oilpressure is linear.

 

[garw]

For the area we are interested in the curve should be linear enough for accuracy In that range .

[ultimate]

Which range do you mean?

[garw]

80-120 deg c Oil temp gauges usually start at 50-60 deg c and go upto 140

[garw]

80-120 deg c Oil temp gauges usually start at 50-60 deg c and go upto 140

[garw]

There is no technical reason why non linear inputs cannot be used I aim to keep the product as simple as possible , so unless there is a real need to Use the full range of the sensor I will keep it simple for now.

[ultimate]

I am with you to keep it simple for the beginning!

 

But I think the oiltemp-gauge will become useless when using a linearity extrapolated from the 80-120deg area.

 

please have a look at the attached image:

In this case you will never see overheating because even at 180deg (10 Ohm) the extrapolated output will show just 135 deg.

 

And for warming up:

70deg (155 Ohm) will be shown as 55

 

Please don't think i will speek your product ill or something. I just want to give you my thinking as an input to optimize it.

Edited by ultimate, 05 April 2017 - 09:01 PM.

[Chris P Duck]

Could you use something like a LTC2997 between the sensor and cluster?