EMS-220 - Pull-Up Resistance

[JGC]

I am new to Dynon products and hoping to get some advice. I am trying to help a friend with his install.

He thinks the pull-up resistors are interfering with the 0 to 5 volts being supplied to the EMS-220 by the 6 EGO gauges that he is connecting.
The only inputs he has available have pull-ups: pins 9, 10, 11, 12, 20 & 21.

Q1. Can the pull-ups on these pins be disabled somehow?

Q2. Why are they included on many of the pins?

Q3. What do they pull up to? I.E. an 5 volts DC buss?

Q4. If I drive these inputs from 0 to 5 volts with a line driver chip - will the EMS-220 input scale 0 to 5 volts correctly?

Regards,
Jon

 

[dynonsupport]

First, What is an EGO gauge?

1) They can only be disabled on 4 of the pins in the system (type C inputs)

2) They are there because most inputs are restive and need a pull up.

3) They pull to 5V DC.

4) The EMS inputs are 0-5V inputs so if you drive to a voltage in this range we can measure it. To do so, you need a driver that can sink current, not just source it.

 

[JGC]

This is very helpful - thank you!
It may be that the EGO gauge's 0 to 5 volt analogue output is being pulled up "toward" the 5 volt DC rail by the resistor.

I will contact the EGO manufacturer to get the specifics on the output circuit,,,, or at least the "drive" capability of the output circuit.

The EGO gauge is an Exhaust Gas Oxygen gauge made by AEM (Advanced Engine Management).

He is using the 30-4110 UEGO.
It is used to monitor "Lambda"

(I tried to include links to the AEM site but was denied) Sorry.

 

[airguy]

Hmmm, this is interesting - and timely. I'm in early Phase I testing on my plane, and I have this oxygen sensor installed and have not been able to make it behave either. I was about to the point of scrapping it entirely but now I think I'll hold off and see what develops here.

 

[mmarien]

I used the PLX AFR controller and had to make a special sensor definition for it. What are you guys using for the sensor definition?

; my sensor definition for the oxygen sensor.
sensor={
     id=AIR FUEL RATIO
     function=PRESSURE
     pins=C37_P8 C37_P22 C37_P31
     name=AFR
     min_val=10.0
     max_val=20.0
     resolution=1
     round=0.1
     is_piecewise=0
     low_coeff={
           c_x4=0
           c_x3=0
           c_x2=0
           c_x1=2.0
           c_x0=10.0
           }
     low_resistance=NONE
     low2high_cross=0
     high_coeff={
           c_x4=0
           c_x3=0
           c_x2=0
           c_x1=0
           c_x0=0
           }
     high_resistance=NONE
     high2low_cross=0
     window=.2
     shift=3
     age=16
     }

AFR gauge at top

 

[JGC]

Background:
The owner is using the AEM Performance Electronics, model 30-4110 Gauge-Type UEGO Controller - Connecting to Dynon EMS-220.
The gauge measures exhaust gas OXYGEN levels. One probe is installed in exhaust stack.

The owner is trying to guess the correct values to put into the Dynon unit to correct/compensate for the pull-ups.
Dynon doesn't provide the equations so it's been several days of trial and error!

I have been looking into ways to counteract the 10k ohm pull-ups that appear to be affecting the readings.
1.
I looked at using two Quad OP amps configured as unity gain amplifiers to drive the input pins of the EMS-220. The problem with this solution is that the OP amps I looked at have the equivalent of approximately one diode drop (about 0.7 volts) before they start sinking current coming into their output pin, so the first 0 to 0.7 volt of gauge output will be wrong.
Also - The OP amp output sinking may change as the OP amp output rises toward +5 volts - which would affect linearity of all of the readings?

2.
A friend suggested applying negative 5 volts DC to the input pin of the EMS-220 thru a 10k ohm precision resistor to "pull-down" the input pin of the EMS-220 to 0.0 Volts DC.
This sounds like a possible solution. Linear Technologies and others make small highly integrated low power +12 volts DC to -5 volts DC power modules.
If the owner is willing, we may try this with a battery and a potentiometer first - to test with -5 volts before building the supply.

You asked about our sensor definition.
I'm not at the hangar right now but will see if I can find out today.

Two things seem strange to me:
First is that the examples in the Dynon documentation have so many significant digits in the factors. They are many orders of magnitude beyond the resolution of the sensors.
Second is that the formulae are not published. It seems that everyone trying to do this would have to send a lot of time playing with the numbers. I'm not a math wizz, maybe that's a factor.

Cheers,
Jon

 

[dynonsupport]

What formula are you looking for? We can't provide a formula to compensate for the pull up unless we know the source impedance of the O2 sensor box. If there was a magic number it would just be built into SkyView.

You can make your life a lot easier if you hook to one of the 4 inputs that has no pull up. This is what everyone else does, but I guess you can't do that if you're trying to read 6 WB02's at once...

I'd think it would be a lot easier to find a rail-to-rail OP amp that can sink current (there are plenty with FET drives that will do this) than it would be to try and pull down. It's a very bad op-amp if it isn't linear on the output with the input, which is kind of the point of op amps.

We support a lot of digits of precision in the formulas because they are coefficients on polynomials and often are very small when dealing with the 3rd and 4th orders.

 

[JGC]

""What formula are you looking for? We can't provide a formula to compensate for the pull up unless we know the source impedance of the O2 sensor box. If there was a magic number it would just be built into SkyView.""

I'm not a math major but the owner is, and he is having a very difficult time figuring out what variables to plug into the the Dynon. He says he can't change them at the Skyview display so he has to go to a PC with a thumb-drive, change the variables, and re-load the Dynon.

""You can make your life a lot easier if you hook to one of the 4 inputs that has no pull up. This is what everyone else does, but I guess you can't do that if you're trying to read 6 WB02's at once...""
Yes, that's correct. This airplane is highly instrumented and all of the enhanced inputs are being used. He is monitoring 6 cylinders to analyse the balance between them.
Since more and more sensors are becoming active, making all input pull-ups programmable to 'none' or adding the pull-ups externally may be a good design revision.

""I'd think it would be a lot easier to find a rail-to-rail OP amp that can sink current (there are plenty with FET drives that will do this) than it would be to try and pull down. It's a very bad op-amp if it isn't linear on the output with the input, which is kind of the point of op amps.""

Yes - I started looking at single rail OP amps for simplicity since we could power them from a regulated supply from the planes B+. Then I discovered the sinking issue - one diode drop. I will look at two rail OP amps - any suggestions of popular ones - ideally quads???

I also thought about opposing the 10k pull-up with a 10k pull-down to a minus 5 volt supply. I think that would 'center' the pin for 0 volts from the gauge BUT I'm not sure how the gauge output circuit would respond???

""We support a lot of digits of precision in the formulas because they are coefficients on polynomials and often are very small when dealing with the 3rd and 4th orders.""

Yes - I was surprised to see so many significant digits since most measuring devices are only accurate to ~~1%.
I assumed that the variables we enter go into one equation that is built into your firmware. We are not writing formulae - we are just entering data into the Dynon?????