Engine Instrumentation Project - PeachParts Mercedes-Benz Forum

winmutt · #1 01-24-2014, 08:49 PM

Quote:

Originally Posted by mach4

The EGT will be running on the Android. Remember, the Android is an old one that is no longer used as a phone, so I'll literally be just using it as a Bluetooth receiver and a presentation screen. My iPhone will be providing phone/navigation/Pandora capabilities as always.

Yes but what adc chip are you going to be using to convert the EGT signal?

mach4 · #2 01-24-2014, 09:21 PM

Quote:

Originally Posted by winmutt

Yes but what adc chip are you going to be using to convert the EGT signal?

I'll be using the MAX31855, probably using the breakout board from AdaFruit, although there are several good ones.

mach4 · #3 01-31-2014, 03:22 PM

Here's a quick project update.

I've been fighting with the bluetooth connectivity for days. This is one of the most ill documented things, not to mention complex I've ever encountered. One would think that it would be a relatively common thing to do data acquisition on an Arduino and pass the data wirelessly to an Android device, but alas it's not. There are lots of examples of controlling an Arduino from an Android by passing single bytes of data, but almost nothing the other way around, particularly using actual data strings.

I've finally got the Arduino passing data strings to the Android relatively reliably. 85 - 90% of the data coming in is good and fortunately I have a way that I can easily recognize the bad data and discard it. If I look at the raw data, it's being received 100% reliably, so there's something with the code that's dropping data. It think it's a timing thing, as what finally got things working was putting in some arbitrary delays. I'm sure that someone out there, will come up with some better code that will make things a lot better.

The way things are currently configured, I'm sending 4 readings per second, which is entirely adequate for temperatures and pressures. If I were attempting to read and present rpms for example this would not be adequate.

Up until now, I wasn't sure if I would be able to get a bluetooth signal from the engine compartment to the passenger compartment. Today I took my breadboard Arduino and stuck it in the front of the engine compartment where I intend to mount it, turned it on and closed the hood. I booted the Android phone into the skeleton application from the normal position in the cabin and was able to read the data stream coming in. The range is not all that great - moving toward the back of the car, I quickly lost the signal, but moving back to the drivers seat provided a very solid signal.

I can now confidently order the rest of the sensors and get the Arduino coded up to supply real data and the Android parsing the incoming data and passing it to the "gauges" part of the program.

More to come...

winmutt · #4 01-31-2014, 04:27 PM

While my experience around the lack of documentation was similar, once it was working it was solid as a rock. What baud are you running at? Did you look at the android bluetooth chat samples? Thats really the easiest way to go. I spent a month dicking around with amarino, boy was that a waste of time. I was pushing full throttle on whatever baud I had it running at (which now escapes me, I remember it being part of the puzzle).

Did you write your own protocol? I did this, picked some random chars for a header and then packed all the data in behind it.

Why would you put your arduino in the engine compartment? It is far better off in the dash somewhere IMO. How are you going to protect it from the elements?

mach4 · #5 01-31-2014, 06:43 PM

Quote:

Originally Posted by winmutt

While my experience around the lack of documentation was similar, once it was working it was solid as a rock. What baud are you running at? Did you look at the android bluetooth chat samples? Thats really the easiest way to go. I spent a month dicking around with amarino, boy was that a waste of time.

Did you write your own protocol? I did this, picked some random chars for a header and then packed all the data in behind it.

Why would you put your arduino in the engine compartment? It is far better off in the dash somewhere IMO. How are you going to protect it from the elements?

I'm using the chat code as a basis. No I didn't write a protocol, just dumping the bytes. The plan is to build a comma delimited string on the Arduino representing the gauge values and then unpack them on the Android side and pass those to the gauge painting routines.

The reason to have the Arduino in the engine compartment is to avoid having to run any wires to the cabin. Also makes wiring up the sensors a whole lot easier. I'll be packaging the Arduino in a sealed relay box (see below) and hook up all the external wires through a standard MB plug. Since the box is plastic the bluetooth module can be mounted inside with the rest of the electronics. The Pro-Mini is shown for scale.

I should only 10 of the 12 plug wires available - 12v power and chassis ground, signal and ground for temp sensors (1-wire interface), two wires for EGT and two signal wires and a shared power and ground for the boost and vacuum sensors - 10 wires.

ROLLGUY · #6 01-31-2014, 07:36 PM

This is an interesting project, so I will be monitoring this thread. Most of the technical jargon is way above my head, but the whole concept is so very interesting, that it should be fun to watch the progress. BTW, I will have the first Trap-Ox bypass pipe finished soon (for those that don't know, I have Paul's fixture). If you need another pipe, let me know. I am also toying with the idea of making everything out of stainless. Sorry for the temporary hijack....Rich

winmutt · #7 01-31-2014, 07:55 PM

Quote:

Originally Posted by mach4

I'm using the chat code as a basis. No I didn't write a protocol, just dumping the bytes. The plan is to build a comma delimited string on the Arduino representing the gauge values and then unpack them on the Android side and pass those to the gauge painting routines.

The reason to have the Arduino in the engine compartment is to avoid having to run any wires to the cabin. Also makes wiring up the sensors a whole lot easier. I'll be packaging the Arduino in a sealed relay box (see below) and hook up all the external wires through a standard MB plug. Since the box is plastic the bluetooth module can be mounted inside with the rest of the electronics. The Pro-Mini is shown for scale.

I should only 10 of the 12 plug wires available - 12v power and chassis ground, signal and ground for temp sensors (1-wire interface), two wires for EGT and two signal wires and a shared power and ground for the boost and vacuum sensors - 10 wires.

Did you fiddle with the baud rates? I seem to remember that causing bad data. I too have a spare bosch box for that very idea but ended up running the wires inside after I abandoned the BT idea. If you see in my first video on the thread I posted, the response time was very dynamic.

With the max6675 I used you had to let it rest between sensor readings. Another thing I banged my head over.

I really like that someone else is running with the BT idea, can't wait to see the final solution. I am also doing water injection as well to keep the EGT's down and will control that my mega. I may end up having to steal some of your ideas

. Wiring is def on of my biggest issues. The cool thing about the android is that you can log stuff, somethin I am sorely missing now.

For intance, I was running 1600F at WOT and recently I adjusted my timing and now my EGT's are far lower, but I don't have any logged datapoints to compare with.