GPS NMEA, Failover Circuit upgrade attempt.

[HellasTechn]

I received the MAX232 chips today. i have put them to use and i get some weird resaults.

At the moment i do not have the GPS receivers to perform the tests but i only have the PC with the termite software to transmit NMEA data to the picaxe through the MAX232 like my last diagram.

I noticed that if PC is connected to GPS2 input (pin 8 of MAX232) the MAX converts the rs232 to TTL and the picaxe can understand the data and work's as desired.

When i change the PC from GPS2 input to GPS1 input (pin 13 of MAX232) and at the moment i attach the connector the MAX chip starts draining much more current and gets hot and gives no output.

After i remove power and allow the caps to discharge it works all right again (without any input connected).

With the PC connected to Both GPS inputs in parallel the problem again shows up.

 

[HellasTechn]

After more testing the max232 i figured out that some chips work for a while and then fail while some others fail from the very first communication attempt. judging by this radical behavior i think that i have got faulty chips.

This time i have ordered the MAX3232CPE that are reccomended by picaxe manual instead of the MAX232 and i will run new tests as soon as i get them.

 

[CDRIVE]

Uh, I think that this thread (now 223 posts long) has seen way too many suspicions of "bad chips". I think you should be looking elsewhere for chip overheating. ... Just say'n!

Chris

 

[HellasTechn]

Please dont take me wrong but i do not understand what you say.

Though rs232 is an ancient communication protocol it is the first time i go that deep useing it so there are many things i dont know about.

 

[CDRIVE]

Hi Constantine. What I was trying to covey is that something is miswired or supply voltages aren't correct. Prior to you receiving the MAX232 chip you had suspicions that there was something wrong with the Picaxe chip you have. I've never had a bad one. Now your new MAX232 is overheating. Sometimes works, sometimes doesn't. This isn't normal. Whether it's an older MAX232 (served successfully for many years) or the latest incarnation (MAX3232CPE) Maxim has released, it shouldn't be getting so hot that it's shutting itself down. Too many of those overheating incidents will shut it down forever. Heat kills semiconductors! That said, I'm a Floridian. So conversely, anything lower than 75 deg F shuts me down too!

Can you post the schematic and some photos? Maybe someone will spot the problem that's causing the overheating. It just isn't normal or something that should be expected simply because the RS232 data isn't what the MAX232 expects to input or output. Data-wise that is!

Chris

 

[HellasTechn]

I understand now.

or supply voltages aren't correct.

That may be the case. I use +5v from the 7805 to pin 16 and GND to pin 15 of the max 232 chip.

I am not sure though if the MAX232 requires +5V and -5V. and i am also not sure if capacitor from pin 2 of the max chip sould be connected with its negative lead to ground.

the circuit diagram i use is exactly as in post #197.

The pcb i made is not the problem because i get the same weird behavior on the breadboard.
I use 1uf capacitors and a 10uf from +5V (pin 16) to GND (pin15) as decoupling capacitor.

Please note that after experimenting with the chips i have come to the conclusion that there is nothing specific that i do that make the chips fail. Some times they fail when i remove an input, other times when i connect an input and some just fail after a few minutes of normal operation.

 

[chopnhack]

I am not sure though if the MAX232 requires +5V and -5V.

Hi K, according to the Texas Instrument data sheet, it just needs a 5V supply : Operates From a Single 5-V Power Supply With 1.0-µF Charge-Pump Capacitors.

I will go back to your schematic, but its important that you have all your caps in place.


Edit: Caps on chip look good to me.
What about power regulation? I don't see any filter caps aside from the 0.1μF caps at the v. regulators, could you be drawing more current on voltage dips? The 12v in I assume is either from marine batteries or generator. The generator can produce swings in line voltage due to engine load demands. Just a thought.

 

[CDRIVE]

I looked back at your schematic and there's an error. C4 on pin2 should be going to GND. You have it connected to the +5V rail. Also, 0.1uF on the output pin of the regulator is too low. 4uF to 10uF is a better choice. You can leave the 0.1uF in there paralleled if you want. It's good practice anyway.

BTW, Chop is correct. The MAX232 only needs GND and a +5V rail. It makes its own -V to produce RS232 (+V -V) level output.

Chris

 

[chopnhack]

C4 on pin2 should be going to GND.

Well spotted Chris!

This is a good reason for why Kris and Hevans frequently stated to draw schematics with a top rail for Vcc and bottom for common, makes it easier to catch errors like that one.

 

[CDRIVE]

I totally agree that placing the + rail on top is good practice and makes it easier to follow. As far a circuit GND is concerned, I almost never draw a common rail, as I prefer individual GND symbols. I realize that some folks like to draw a common rail with a single GND symbol connected to it. Sometimes, (when that's convenient) I'll do that but only with simple circuits.

On another note, I peeked at the MAX232 datasheet. The Vcc current (all outputs open) should not exceed 10mA. Definitely not enough current to heat it much.

Chris

 

[HellasTechn]

Hi K, according to the Texas Instrument data sheet, it just needs a 5V supply : Operates From a Single 5-V Power

Got that.

What about power regulation? I don't see any filter caps aside from the 0.1μF caps at the v. regulators, could you be drawing more current on voltage dips? The 12v in I assume is either from marine batteries or generator. The generator can produce swings in line voltage due to engine load demands. Just a thought.

Power comes from a P.C. powewr supply (meaning it is regulated and capable of giving out several amps).
About regulation i will add a 10uf decoupling capacitor to picaxe from pin 1 to gnd.

Also, 0.1uF on the output pin of the regulator is too low. 4uF to 10uF is a better choice.

On the datasheet they have both 0.1uf and 10uf in parallel. I will correct that.

I looked back at your schematic and there's an error. C4 on pin2 should be going to GND.

The reason i connected it this way is because of this comment i found online by someone:

"The use of the capacitor on pin 2 shows a slight deviation from the "Typical Operating Circuit" given in the Texas Instruments datasheet (Figure 4). In the datasheet, the pin is capacitively coupled to ground through a 1uF capacitor. Here you have it coupled to VCC. That does not matter much, because both VCC and GND are AC ground. However, this configuration is not a mistake. It appears in the original Maxim datasheets for this device!"

I will change the capacitor and place it with positive side to pin2 and negative to gnd. All that on the breadboard and i will report the resaults here asap.

THANK YOU BIOTH YANNIS AND CHRIS !

 

[CDRIVE]

Also make certain that Pin6 (-V) has the 1uF cap polarized properly. I know your schematic indicates Neg to pin6 and Pos to GND, which is correct but double check this.

Chris

 

[HellasTechn]

Yes i have checked it and i can confirm that Negative lead goes to pin6 and positive goes to pin15 GND.

Today i was engaged with some other work so i had no time but in the morning i will run the circuit with a new MAX232 chip.

Good thing i bought 10 of them because so far 5 have kiked the bucket.

 

[CDRIVE]

I don't want to see you smoke more of them so double check the polarity of the other caps too. I've been staring at your schematic to the point of bleary eyes but I just can't see what's popping the MAX. What's the voltage rating of those caps? Have you Ohm'd them out to insure none are shorted? .... Grabbing at straws!

I sure wish you had a current limited supply to do your testing with.

Chris

 

[HellasTechn]

I don't want to see you smoke more of them

Too late

so double check the polarity of the other caps too.

I have checked them all at least 5 times and all are correct.

I've been staring at your schematic to the point of bleary eyes but I just can't see what's popping the MAX.

Yes it just does not make sense.

What's the voltage rating of those caps? Have you Ohm'd them out to insure none are shorted?

The caps are 1uf 63V and all ove them i have tested with an ESR meter and give me an ESR of about 0.6 and Vloss 1% so the caps are is very good condition.

I sure wish you had a current limited supply to do your testing with.

It sure is an extremely useful instrument but no i dont have one.

After altering the circuit as you suggested and making it exactly as shown in the datasheet i have run more tests.
The MAX232 was running perfectly untill the time i started pluging and unpluging the GPS inputs...

 

[HellasTechn]

While the MAX was still running ok i tried to use the code with different frequencys. All frequencys worked but 4 MHZ.

After removing the MAX chip and testing again with the resistors the picaxe worked on all frequencys but 4 MHZ.

So my first thought that the picaxe at 4Mhz might just not be powerfull or fast enough is gaining ground in my head.

 

[hevans1944]

Has anyone checked for ground loops and common-mode voltages between the serial port on the PC (or a USB-to-serial converter) and the MAX chip? It sounds like something is overloading the MAX inputs, perhaps by a LOT to make it get hot and fail.

I just now stumbled onto this thread but it appears you are trying to decipher GPS "sentences" with a PICAxe microcontroller... to what end I don't know, but I will go back and read the thread over the next few days to see what the final objective is. Sounds like something I might be interested in exploring since GPS receivers are getting to be dirt-cheap for hobbyist use, and I know virtually nothing about them!

I recently had an Elevation Certificate issued by a surveying company in Florida for a nice bit of property we are trying to purchase that happens to lie on a flood plane. The property is at the edge of, and totally enclosed by, a Zone AE flood zone. I would like to set a GPS receiver and measure elevations on the property after integrating the GPS results over a period of several days to improve the elevation resolution and (hopefully) accuracy and then compare what I measure to the results reported by the survey team. I know there is a procedure to do this for latitude and longitude measurements. so I am hoping it also applies to altitude.

There is a huge amount of theory and math involved in geodetic surveys and I am just beginning to learn about it. Even estimating Base Flood Elevations (BFE) is fraught with complications related to calculating water flow. And no doubt "global climate change" will affect things too, since the vertical datums are based on mean sea levels measured at various control points as well as GPS data. I admit I have a lot to learn, but hopefully I will have time to learn it in the warm Florida sunshine...

 

[HellasTechn]

Has anyone checked for ground loops and common-mode voltages between the serial port on the PC (or a USB-to-serial converter) and the MAX chip? It sounds like something is overloading the MAX inputs, perhaps by a LOT to make it get hot and fail.

They are actually being tested useing a PC serial port. In the future it will be used with the GPS serial port.
To be honest, no, and i have no idea how to do so.

I just now stumbled onto this thread but it appears you are trying to decipher GPS "sentences" with a PICAxe microcontroller... to what end I don't know, but I will go back and read the thread over the next few days to see what the final objective is.

If you are patient enough to read through all the posts then on #1 post in this topic there is a link to an other topic i have started about 3 years ago. If you read it all the way down to this post you will get whole the picture.

sounds like something I might be interested in exploring since GPS receivers are getting to be dirt-cheap for hobbyist use, and I know virtually nothing about them!

This website is a very good place to start: http://www.gpsinformation.org/dale/nmea.htm

Also:

I recently had an Elevation Certificate issued by a surveying company in Florida for a nice bit of property we are trying to purchase that happens to lie on a flood plane. The property is at the edge of, and totally enclosed by, a Zone AE flood zone. I would like to set a GPS receiver and measure elevations on the property after integrating the GPS results over a period of several days to improve the elevation resolution and (hopefully) accuracy and then compare what I measure to the results reported by the survey team. I know there is a procedure to do this for latitude and longitude measurements. so I am hoping it also applies to altitude.

If i understand correctly what you need to do is to setup a gps receiver on a specific point and then log the altitude indications through a period of time and then analyse and compare them. Is that correct ? I am sure we can write a code for a picaxe to log altitude over time.

 

[HellasTechn]

Allright, according to this website http://www.gpsinformation.org/dale/nmea.htm we need GGA sentence
GGA = essential fix data which provide 3D location and accuracy data. So like i said before we* should be able to write a code to do what you need.

* That we goes for you ash and me also to get to learn.

 

[CDRIVE]

USB/Serial converters commonly output ±5V. The MAX232 inputs can handle up to ±30V. Early RS232 standards found data levels as high as ±30V.

I'm still lost!
Chris