:roll: Honestly… I beat at a problem for a while before I ask for help… but I'm stumped as to what I am doing wrong here.

The ADS1115 Datasheet specifically tells you what to put in to get a reading… and I think I'm doing just that. The problem is… I'm not getting any errors, but the chip constantly reports MSB = 0 and LSB = 0 and it does not matter if I tie the AN0 pin to 3.3vdc or Ground.

I have three outputs… each is a Textbox and my routine sends the data there. A TIMER1 pings the routine once a second.
I also have text printing to the debugger area so I know it is running the routine. I have tried putting the data in as binary <COLOR color="#BF0000">&b00000000&</COLOR> as HEX <COLOR color="#BF0000">&H00&</COLOR> and as a plain old <COLOR color="#4000BF">decimal</COLOR>. Same thing.

here is the code… maybe I'm just too close… overlooking a blatant error.


I have an I2C 8-Bit output chip… maybe I'll install it and try lighting up some LEDs. I've done I2C before with both chips in PYTHON… but that was a different command structure. I really like PIGPIO and want to get this to work… all my other functions INPUT/OUTPUT/PWM/SERVO all work great.

Thanks,
Jerry

Askjerry said

…Datasheet specifically tells you what to put in to get a reading… and I think I'm doing just that. The problem is… I'm not getting any errors…

Well, you might be getting errors that you are not checking for.

I'm probably being an old woman, but I'd check the return values for all the write commands you are using…just in case.
Example:-

I swear this chip is going to drive me crazy… a short drive actually.

After not getting the ADS1115 to work properly I decided to try another IC that I am familiar with, the PCF8574 Extender Chip.
The datasheet is here:

Page 9 shows the address as follows: 0100aaar where aaa is the address set by jumpers, and r is 0 to 1 depending on if you want to write (0) or read (1).  As my chip address pins are wired to the ground, it would make the address to be 0100 0000 and I would have expected the number to be left to right &H20&. However, that failed… it was actually &H02& as you have to read the two numbers left to right with the bits being 1248.

Ok… I got that to work just fine. I was able to toggle some LEDs on and off proving the i2c commands work.
Here is a code snippet.


Taking what I just learned… I wanted to apply it to the ADS1115 chip as it is necessary for the project I am working on to read multiple sensors.

The PDF for the ADS1115:

Jumping into page 35, they give you the proper sequence of codes to set the unit up for Port 1, Address 0, Continuous mode, 3.5v which is exactly what I need for the first sensor.

1.Writeto Config register:
  –  First byte (first7-bit I2C address followed by a low R/W bit)
  –  Second byte (points to Config register)
  –  Third byte (MSB of the Config register to be written)
  –  Fourth byte (LSB of the Config register to be written)
2.Write to Address Pointer register:
  –  First byte (first7-bit I2C address followed by a low R/W bit)
  –  Second byte (points to Conversion register)
3.Read Conversion register:
  –  First byte (first7-bit I2C address followed by a high R/W bit)
  –  Second byte: the ADS111x response with the MSB of the Conversion register
  –  Third byte: the ADS111x response with the LSB of the Conversion register

So we look at the numbers… and convert them the same way.
1001 0000 → 09
0000 0001 → 80
1000 0100 → 21
1000 0011 → C1

---

1001 0000 → 09
0000 0000 → 00

---

1001 0001 → 89

The code snippet works out to
    
And the entire program being:


Except… the code doesn't work. The PCF8574 is just fine… no issues at all.
The ADS1115 Still throwing errors. I don't know why.

Sections 1 to 5 are fine.
Sections 6 to 9 fail.

I'm at the point where if someone wants to take this on, I'll send you an ADS1115 module so you can figure what I am doing wrong.
I need this to read the airflow sensor, load cells, and temperature sensors. I just don't get why it is not working.

Current ADS1115 program zipped.

Askjerry said

I swear this chip is going to drive me crazy… a short drive actually.

…The PDF for the ADS1115:

Hi Jerry, I'll try to find time to look at the ADS1115 info later today.

But in the meantime, I don't understand why you are only checking the function return value (intReply) after the case statement.
This is fine for Case 1-3 & 5, but not for 6-8 where intReply will only show the return value for the last of the function calls.

I guess if you wanted to be economical with your source code you could do something like this:-

…but of course this will make the Fail Code description meaningless:-

 It was basically just to see IF there were any failures, and if so to locate what section of the code so that I could then make more lines to specifically go into each CASE statement and drill it down. If you want me to ship you a module please let me know. It has to be something simple I'm missing… the extender chip works flawlessly, and when I was doing python programming the library function was flawless. (It operated differently however, you gave it the address and it did everything else… so I don't know all the internal workings, unfortunately.)

I need to get this working with Gambas3 because it allows me to make an incredible GUI when compared to other languages.

thanks,
Jerry

Askjerry said

…The datasheet is here:

Page 9 shows the address as follows: 0100aaar where aaa is the address set by jumpers, and r is 0 to 1 depending on if you want to write (0) or read (1).  As my chip address pins are wired to the ground, it would make the address to be 0100 0000 and I would have expected the number to be left to right &H20&. However, that failed… it was actually &H02& as you have to read the two numbers left to right with the bits being 1248.

Ok… I got that to work just fine…

I'm sorry. I spent a couple of hours this evening looking at this, but I don't really understand how this bit works with the PCF8574.

The datasheet shows the most significant bit of a byte being sent first. But as this complies with the I2C standard, I would expect the pigpio library to take care of this, e.g. you give the write function the required address and the function sends the bits in the required order; msb to lsb

However, you have proved your code works with all bytes reversed, so I can't argue with that.

My only suggestion is to look at the Python code that you have used successfully on the ADS1115, and then try to replicate this in Gambas,

 That's not going to work… it has something like READ_ADS(0) to read the port 0… it doesn't get into specifics as to how it does it.

I can get the extender chip to work… but not the ADS1115 chip.

Seriously… to anyone reading this… I'll send you a module if you want to give it a shot… I need to get this working and don't know where else to turn.

Jerry

Askjerry said

That's not going to work… it has something like READ_ADS(0) to read the port 0… it doesn't get into specifics as to how it does it…

Your Python code (with lines like READ_ADS(0)) will reference libraries. What I am suggesting is that we look at the library code.
This library will probably also be written in Python (or it may be a C library) and should be easy to spot in the list of import declarations.

Thanks,

I will go through my archive SD card (I completely started over with this Raspberry Pi) and I will see if I can find and deconstruct the code.
(That would be nice if I can pull it off.)

But the offer is still out there if someone wants to PM me I'll send an ADS1115 module to you so you can try to figure it out.

Jerry

Hi… I am attempting to compare the speed of how numerous information tests can i get from ADS1115 between python and C++. So distant i have tried the python utilizing Adafruits library. Now i am searching for a test code in C++ to require a perusing but i might not discover an awfully fundamental one. I have seen that the wiringPi library includes a ads1115.h, but i seem not discover documentation of a test code.
Turnkey PCB Assembly - Bittele Electronics

BriggTrim ->

No, there is no documentation on how to use this chip with Gambas and PIGPIO. Once I figure it out, or someone else does and tells me how… I will post the results and code here.

Jerry

 Still no answer here on reading the ADS1115…

I realize that the PIGPIO can also be used with C and other languages but the calls will remain the same. Does anyone here have code in another language that reads the ADS1115 using PIGPIO? If so, please post your code… I may be able to use what you have to successfully read the chip from within GAMBAS. If not… I may be able to call a routine in another language if needed.

Thanks,
Jerry

 Still no reply… wow… I really want to get this project moving.
It looks like my option is to program an Atmel AVR in BASCOM and then read it via SPI… that is sort of Rube Goldberg… and I didn't want to interface one chip with another… because my subscribers are likely not going to have BASCOM, and they will end up using AVRDUDE to program one chip to read another… uggg.

Python has Adifruit routines that work… is it possible to make a call to PYTHON???

I remember in the old TRS-80 days… yeah, I'm freaking old… we could write a BASIC program and put an ASM call to pull an assembly language routine in. Is there a way to pull in the Adifruit and/or Python routine?

I REALLY need to read this ADS1115 chip.

Jerry

I REALLY need to read this ADS1115 chip.

I am not really qualified to answer this, as I don't even know what a sADS1115 chip is. However, my thoughts are that any library for Python is probably written in C or C#. If this is the case, then Gambas should be able to access it. The Gambas help for this is here.

With the help of Guido Lutterbach I am about 85% able to get this chip to work.
He sent me a python code which works and uses no fancy library to function.
ADS1115Runner.py — smartyPies
I was able to understand enough of it that I am now getting changing values when I turn the potentiometer.

There is something called Big Endian and Little Endian which basically means that for some reason if you want to send ABCD sometimes you need to send CDAB because of how the Raspberry Pi handles things internally.
(It makes no sense to me why this happens… but it does.)

To accomplish this feat, he uses this line of code in Python…   value = (value>>8 | value<<8) &=0xFFFF

I found this reference… but I don't understand what it is trying to tell me.
https://gambaswiki.org/wiki/comp/gb/stream/byteorder

Can someone here tell me what I need to do to accomplish this in Gambas?
(Translate the above into a command or short routine in Gambas.)

He is also doing some sort of 2's complement function to take the returned value and make it into a readable decimal… those are…

LEtoBE

BEtoLE

Eventually I will learn enough python to fully understand this… but for the moment… if someone can translate… that would be freaking awesome.

Guido breaks it all down here… I'm just slowly getting it piece by painful piece.
http://smartypies.com/projects/ads1115-with-raspberrypi-and-python/

I did learn that I was using the wrong method to send the configurations… not realizing that the i2cWriteWordData( x , y , z) was NOT asking for the chip address again… I feel so silly now… but instead asking for:

 <COLOR color="#0000FF">i2cWriteWordData( Currently Open Bus , REGISTER TO WRITE , WORD TO GET WRITTEN)</COLOR>

So for now, I am going through the datasheet… writing down the BITS to make the CONFIGURATION, then manually swapping the BYTES and putting it into the code.

The numbers coming out are changing… and sometimes gibberish… but at least I know now that I need to post-process before dumping them into a TEXTBOX.

I think I found at least a partial answer… a command called <COLOR color="#0000FF">CShort (Expr AS Variant)</COLOR> getting closer at least… then eventually I'll replace this thread with an exact step-by-step to help out anyone else from going through this!

Thanks in advance!
Jerry

<COLOR color="#40BF00">AFTER 5 MONTHS I GOT IT!!!!</COLOR>


On a 3.3v setup I am getting readings from 2 to 26,500 which means I'm getting somewhere around 0.000125v accuracy. (Theoretical)
Probably way more than I need to read a wind gauge… but hey… IT WORKS!!!!

<COLOR color="#BF0000">WHOOO!</COLOR>

I wanted to do an update… calculating the HEX numbers to send to the chip has been a pain in the posterior… so I made a helpful spreadsheet to figure them out… it does all the work for you.

For an example, I wanted to configure the chip to use Port 0 to Port 1 in differential mode, then connected Port 1 to ground, and Port 0 to the center tap of a potentiometer… the result was &h8383 as shown in the spreadsheet. You simply fill out the multiple questions in the lower left based on the chart… and it provides the programming number for the configuration.  

Sample GAMBAS QT file

Then I created a simple demo program with comments to get you started.
Let me know if it is helpful to you!

Spreadsheet to calculate values

Unzip the above and copy the folder to wherever you keep your GAMBAS projects.


Thanks,
Jerry