Order by
Group by
Topic

Looking for all posts by mhardy

Page 1 of 3 out of 57 messages.

RE:G120 E Analog inputs not working by mhardy on Feb 8, 2017 at 3:27pm

Was this resolved?

I have seen similar things with the G120.
It is totally intermittent and seems like a power/G120 boot-up race condition or something.

Anyway I will power-up the device, and only sometimes do my analogs read 0.
To fix this I can reboot the G120 (not power cycle) and the analogs come back.

Any idea on how to fix?

NETMF 2016 R1 (4.3.8.1) COM2 (RTS & CTS) .Handshake.RequestToSend by mhardy on Oct 19, 2016 at 8:15am

Com2 handshaking appears not to "yet" work?
I see in some places in the forum where there was a problem with handshaking and in other parts of the forum where it has been fixed.

I am noticing when transferring larger (2k +) byte[] buffers that the data gets screwed up. I don't believe I saw this same problem with an earlier release.

Can anybody confirm that the com2 serial port on the g120 should handshake correctly, i.e.
.......setting .Handshake = System.IO.Ports.Handshake.RequestToSend will control data flow over this serial port.

RE:$10,000 for contract developer help.... by mhardy on Aug 27, 2016 at 7:01am

I agree.
That's why this was my first stop for help with my problem.

Gus and all, anybody whom can help, we are willing to compensate....a little, not like we are Oceans 13 or something!

$10,000 for contract developer help.... by mhardy on Aug 27, 2016 at 6:42am

I am not sure if this is the proper way of going about it.
GHI let me know.........
Would be happy to hear from your engineers also. You build a great product. I hope this is the best way to go about it.......Many smart peeps doing similar things!

We have a board based around the g120.
We have a need to code for a industry standard protocol which was devised for bigger metal.
We have the code talking the protocol via standard .Net Framework.
We have ported most of it down to .NetMF.....
The road block is xml serialization.

The protocol was built by a group, and with a group it does everything for everyone, but no one in particular.
The protocol is built with a pile of metadata xml rules checking and such, which I gave up on trying to port to .NetMF...about a year ago.
We are talking classes inherited from classes, deep arrays and such.
Thought I would take another shot, as the protocol is starting to take more of foothold.

We need a developer that may have gone down this path and has the answers.....that I never had time nor the inclination to delve deeper.

This would be a contract job with steps of viability along the way to get paid $2,500 at a shot.
The total of $10,000 would be presented at the time we were assured of being able to go for certification.

We also have many more technological challenges to face, which success on project #1 above would make you candidate #1 for future contract endeavors.

Sincerely....this is for real.
Michael Hardy
Control Dynamics

BTW: Our board builder is Silicon Mountain....Nampa, ID

RE:G120 Ram Usage by mhardy on Apr 9, 2015 at 2:09pm

Ok...great.

Another data point on understanding the characteristics of the G120

G120 Ram Usage by mhardy on Apr 9, 2015 at 12:50pm

I was going about adding more code for a G120 custom board.
I was about to add some more references for things like xml, http, cryptogrophy, etc.
I figured I would check how much memory was left, so added code below to show.

uint freemem = Microsoft.SPOT.Debug.GC(true);
Debug.Print("Free Ram = " + freemem);

Free Ram = ~6MB

Wow, I thought, how could my code eat up that much memory...I realize netMF + GHI takes some room also.

So, I loaded a program (shown lower down) that simply just prints ram available.
Ran this program and:

Free Ram =~7MB

Question: Does netMF + GHI assemblies require 7MB of RAM (at minimum)?

Streamreader.Readline() all of a sudden stops working? by mhardy on Mar 24, 2015 at 10:55am

I have been using the same code for weeks to read a file from sd card, via Streamreader.Readline(), this morning it stopped workin for no apparent reason.

Checked the file I was reading....nothing apparently wrong.

Ended up implementing ReadLineEX() code, link below.
This got me back running!!

Anybody have the same experience of .netMF all of a sudden not working as it use to?



https://www.ghielectronics.com/community/forum/topic?id=8577&page=1

RE:G120 Analog Input Range by mhardy on Mar 19, 2015 at 9:10am

Thanks for the verification and quick update to the user manual...

FYI,
The manual still has a misprint stating that analogs are 10-Bit, however.

Section:
1.3.G120 Module Key Features
.
.
.
.
8 10-Bit Analog Input

RE:G120 Analog Input Range by mhardy on Mar 18, 2015 at 6:40pm

Yes - just looking for confirmation from GHI.

The NXP LPC1788 User manual states:

12-bit Analog-to-Digital Converter (ADC) with input multiplexing among eight pins,
conversion rates up to 400 kHz, and multiple result registers.
The 12-bit ADC can be used with the GPDMA controller.
-
10-bit Digital-to-Analog Converter (DAC) with dedicated conversion timer and DMA
support

RE:G120 Analog Input4, P1_30 (Pin 59) by mhardy on Mar 18, 2015 at 6:22pm

Thanks Aron,

I also grounded P1.30 and my app runs just fine!

I also connected via fez config (via usb) and was able to erase the app, update firmware and do other functions, that I have done in the past.

The only function that appears to not work is updating TinyBooter via fez config, and I would assume via the command line also...

So, I guess all I need to do is add a pogo pin to disconnect P1.30 from any circuitry in the event that I need to update TinyBooter.

RE:G120 Analog Input Range by mhardy on Mar 18, 2015 at 4:57pm

Yes Brett... I fully understand how analogs work!
My circuitry converts current to volts, where 20mA = 3.3V.
At 3.3V on the analog pin, I am seeing a count = 4095.

The manual says the analogs are 10bit (0-1023)....but I am seeing 4095 (12bit).

I am asking for verification that the manual is incorrect?

G120 Analog Input Range by mhardy on Mar 18, 2015 at 3:35pm

The user manual (March 9, 2015) say's the G120 has 10 bit analogs (0-1023).

AnalogInput.ReadRaw() is returning values from 0-4095, indicating the G120 has 12 bit analogs.

I assume the user manual is incorrect, right?

btw: My custom board has circuitry connecting the analog inputs to sensors that are 4-20mA, when I max the sensor out (20mA) I am seeing the digital analog count value = 4095

RE:G120 Analog Input4, P1_30 (Pin 59) by mhardy on Mar 18, 2015 at 3:27pm

Gus -

Have you had a chance to double check this yet?

RE:G120 Analog Input4, P1_30 (Pin 59) by mhardy on Mar 17, 2015 at 6:42pm

Gus -

Yes, please double-check!

It is critical that we know EXACTLY what to expect.

If we do not understand this correctly, it could totally Brick boards that we will send into the field.

RE:G120 Analog Input4, P1_30 (Pin 59) by mhardy on Mar 17, 2015 at 6:28pm

So, if P1_30 is grounded on power-up, my application will execute, all AOK, correct?

Only if I want to update TinyBooter, for some reason, will I need to make sure that P1_30 is NOT grounded, correct?

RE:G120 Analog Input4, P1_30 (Pin 59) by mhardy on Mar 17, 2015 at 6:17pm

Gus -

Prefer to not add circuitry......

Not sure I understand what you mean in your statement:
"Also, this effects the GHI loader, not the firmware. You can add a jumper if you ever needed the GHI loader after production."

My understanding is that, the GHI bootloader is ALWAYS required...production..debug..whatever?

The documentation (the G120 SOM user manual) say's:
GHI bootloader...blah...blah & executes TinyBooter
TinyBooter..blah..blah loads the firmware.
The firmware executes my application.

Perhaps you are saying that the P1_30 pin cannot be grounded on power-up if one plans on sending commands to the GHI boot loader?

G120 Analog Input4, P1_30 (Pin 59) by mhardy on Mar 17, 2015 at 4:15pm

Received my custom board based on G120 SOM last Friday.

Turns out the 600VAC input circuitry outputs an analog value, instead of a digital value ...oops. So, I need to hook this circuit to an analog input instead of a digital input as I had initially planned.

Anyway I have used up all G120 Analogs with the exception of Analog 4, P1_30 (Pin 59).
I avoided this input on initial design because of user manual note below:

* Pin 59 P1_30 AD4/COM4 OE must NOT be grounded on power-up for GHI Bootloader to function correctly.

Does this mean AI4 is totally worthless?
Perhaps it means as long as the input is floating (could be floating at 0V) and not wired to ground everything works fine Huh

Anybody know "EXACTLY" what the note is saying?

Anybody use Analog 4....any problems?

RE:Scary G120 (Cobra II eco) NetMF, System.Text.UTF8Encoding, System … by mhardy on Dec 30, 2014 at 5:51pm

John -

Received new Cobra II (eco) this morning and have been running it since.

Works as expected, i.e. no lockup or spewing of crash reports. At a count of ~ 20,000,000 cycles now. On old board I was lucky to get to a count of ~30,000 or so.

Surprising, each loop takes about 1ms to complete.........

RE:Scary G120 (Cobra II eco) NetMF, System.Text.UTF8Encoding, System … by mhardy on Dec 22, 2014 at 3:42pm

Reinhard,

Just tried what you said....Worked just as you said....thanks

RE:Scary G120 (Cobra II eco) NetMF, System.Text.UTF8Encoding, System … by mhardy on Dec 22, 2014 at 3:40pm

OK, I will place an order.

The RMA is not that important to me. Finding the root cause is.
When you get an RMA, can it be marked for failure analysis?

Page 1 of 3 out of 57 messages.