well the dongle made a 90kb file when I first plugged it in, the blue light is flashing, but nothing else is happening.
I'll try replugging it now it has a fix.
Nothing obvious, think I'll have some lunch and try again later.

on the battery thing, the dongle is enough to keep my powerbank on. I know the Pi has turned off because it's no longer on samba. but the dongle is still powered holding the battery on.

well the dongle appears to be working now it's up to 92kb and counting. But still nothing from the module.
Going back to the board now

Select to expand quote

decrepit said..
on the battery thing, the dongle is enough to keep my powerbank on. I know the Pi has turned off because it's no longer on samba. but the dongle is still powered holding the battery on.

Well that's weird, even though the Pi is off and the dongle out, the battery stays up. It only turns off when I disconnect the Pi. I'm sure the battery turned off when I shut down the Pi manually.

Select to expand quote

boardsurfr said..
Update with some info about using USB battery packs at boardsurfr.blogspot.com/2018/05/almost.html

Peter how about a visible LED to let you know the Pi is running?

So that's what a flaka is, you learn something every day.

Don't know what's happening with the module now, but it's stopped saving the baud rate, reverted back to 9600 overnight, and now won't hold it through a brief disconnect reconnect. I'm sure I'm getting it right now. save PRT, save receiver, save CFG.
Thought maybe I'd knocked the battery loose but that looks OK, I'll put the meter on it tomorrow and check.

Select to expand quote

decrepit said..
Don't know what's happening with the module now, but it's stopped saving the baud rate, reverted back to 9600 overnight, and now won't hold it through a brief disconnect reconnect. I'm sure I'm getting it right now. save PRT, save receiver, save CFG.
Thought maybe I'd knocked the battery loose but that looks OK, I'll put the meter on it tomorrow and check.

I think some designs have a capacitor instead of a battery, so they will withstand a short power outage, but after that it will revert to the stored defaults.

One way to test if you are saving things is to change the update frequency. If you go into RATE(rates) and change the measurement period to 500ms versus the default, which I think is 1000ms, you will get two updates a second. Alternatively, go the other way and change it to 3000ms and you will get one update every 3 seconds. If these settings are also lost, then something is wrong with the battery.

Make sure you are seeing valid data in the text view and packet view after you change the port and receiver baudrates, in order to verify that the baudrates are correct. Then do the save from the CFG tab.

Not seeing what's happening with the Pi and the dongle is ok when everything works fine, but it's obviously no good when trying out new things. I've gotten the 2.13 in e-ink HAT from Waveshare (https://www.amazon.com/gp/product/B071S8HT76/) to work. So far, it only shows status messages, not speed, but that will come. Not sure if you can get it for a reasonable price in Australia. Also, it fits best to a Pi Zero W (the one with the header).

Select to expand quote

decrepit said..

Well that's weird, even though the Pi is off and the dongle out, the battery stays up. It only turns off when I disconnect the Pi. I'm sure the battery turned off when I shut down the Pi manually.

What battery are you using?

Select to expand quote

boardsurfr said..

What battery are you using?

it's a 4000mah solar rechargeable job.
www.jaycar.com.au/4000mah-solar-rechargeable-power-bank/p/MB3723

I did read the switch off current somewhere, can't see it now, but I feel it was less than 100mA

Select to expand quote

FormulaNova said..
I think some designs have a capacitor instead of a battery, so they will withstand a short power outage, but after that it will revert to the stored defaults.

Definitely a battery that's what it looks like and it's what the schematic shows.

It didn't hold the rates change, but I think I better go a bit more extreme to make sure it's actually happening.

Select to expand quote

decrepit said..
Don't know what's happening with the module now, but it's stopped saving the baud rate, reverted back to 9600 overnight, and now won't hold it through a brief disconnect reconnect. I'm sure I'm getting it right now. save PRT, save receiver, save CFG.
Thought maybe I'd knocked the battery loose but that looks OK, I'll put the meter on it tomorrow and check.

Sounds like the battery may be empty (if you're talking about the NEO-8 module). IIRC, the "real" NEO-M8N modules are supposed to have Flash memory to which they can save settings without needing a battery.

Can you enable GLONASS and GALILEO at the same time as GPS?

Select to expand quote

boardsurfr said..

Sounds like the battery may be empty (if you're talking about the NEO-8 module). IIRC, the "real" NEO-M8N modules are supposed to have Flash memory to which they can save settings without needing a battery.

Can you enable GLONASS and GALILEO at the same time as GPS?

Yep, I changed the rate to 5s, can really tell that that is what is happening, but disconnect and plug back in and its 1s.

This model has a battery connected directly to pin 22, labeled "V_BCKP"




I guess you can't tell by the label, but there's the battery top left.

And yes it has all sorts of formats I've never heard of, including GPS and Glonass, not sure about Galileo, but I think so.

I looked at the GPS module that I have, the exact same type, and there is indeed a battery. It measures at 1.028 volts, which is a tad low, although mine seems to retain its config. The battery can be measure by using a multimeter on the top surface and the corner on the pcb where its mounted.

(Edit: Maybe its not a battery after all. I ran the GPS up and after disconnecting the voltage on the 'battery' it is now around 2.4v. It could be a capacitor. There are certainly diagrams online that show a capacitor on the V_BCKP pin. Decrepit, your module looks slightly different in that the metal tag that holds down the battery/capacitor is aligned differently. There are three triangular tabs under the battery and the tab on mine is soldered to the one furtherest away from the gps chip closest to the UART connection point. You can measure the voltage on the one that points towards the ublox chip or the one that is on the outside of the board. Yours looks like it is soldered at that tab that is pointing to the ublox chip. I wonder if this is a manufacturing fault and its soldered to the wrong place?)

In the config CFG screen of U-centre, it automatically highlights the 0-BBR and 1-Flash entries when you go to save the config, so it implies that it saves to flash as well. I don't know when it uses the flash config versus the ram memory.

Maybe these boards are old stock and the batteries are on the way out? The same board looks to support previous generations of GPS chip, so maybe that's the case?

I think if writing scripts for these it would be good to initialise them every time with the default baudrate and then change the baudrate straightaway, just in case the config is lost or a new module is used.

Select to expand quote

FormulaNova said..
Maybe these boards are old stock and the batteries are on the way out? The same board looks to support previous generations of GPS chip, so maybe that's the case?

Sound quite likely.

Select to expand quote

FormulaNova said..
I think if writing scripts for these it would be good to initialise them every time with the default baudrate and then change the baudrate straightaway, just in case the config is lost or a new module is used.

Initialization from the Pi seems to require the correct baud rate for some setups (but apparently not for the VK172 dongle), so it's a bit of a catch-22. But by trying to read from the dongle at different rates, it should be possible to find the current baud rate, then change it, and verify. I'll put that on my to-do list. If you don't want to wait, I found a python script that does that at:
github.com/somu1795/baudrate

Select to expand quote

decrepit said..
Yep, I changed the rate to 5s, can really tell that that is what is happening, but disconnect and plug back in and its 1s.

Sounds like the battery is dead, and the module does not have flash memory. Not all ublox 8 modules have flash. Your module is supposed to be a NEO M8N, which would have flash. But some fake modules don't, seems like your module is one of them. It may still have an original ublox 8 chip inside, but it's apparently not a genuine NEO M8N. Not a big surprise, since ublox sells the NEO M8N at about $12 (for quantities of 250+), and that does not include the board the chip is mounted on.

You may want to see if you can replace the battery with a new one. If the chip "forgets" everything every time you unplug it, that has some other consequences, too. For example, you can't have a "hot start", which is faster. More annoying is that the reported time will probably jump by a few seconds when the chip finally gets the correct number of leap seconds, which can take 10-15 minutes.

BTW - I probably was wrong thinking that seeing only 32 channels in ucenter is a sign of a fake ublox 8, according to this: portal.u-blox.com/s/

I got some windsurfing data today with the Pi Zero and the GPYes dongle. The good news is that is worked ok. The bad news is that the accuracy was not as good as expected, with typical errors of +- 0.8 knots; I expected below 0.5. So I did some playing around with different positions of the GPS unit relative to the Pi; here's a plot of the accuracy estimates for 4 different configurations:


For the first ~300 points, the GPS was about a meter away from the Pi (on a USB extension cord). For the second 300 points, it was right next to it; then right on top of the e-paper display; and finally on the battery, next to the Pi and a bit below the display. The average error estimate was about 3 x higher for the last configuration - which is unfortunate, because that's what I used. So the best positioning of the GPS will need some more research.

Select to expand quote

boardsurfr said..

FormulaNova said..
Maybe these boards are old stock and the batteries are on the way out? The same board looks to support previous generations of GPS chip, so maybe that's the case?

Sound quite likely.

FormulaNova said..
I think if writing scripts for these it would be good to initialise them every time with the default baudrate and then change the baudrate straightaway, just in case the config is lost or a new module is used.

Initialization from the Pi seems to require the correct baud rate for some setups (but apparently not for the VK172 dongle), so it's a bit of a catch-22. But by trying to read from the dongle at different rates, it should be possible to find the current baud rate, then change it, and verify. I'll put that on my to-do list. If you don't want to wait, I found a python script that does that at:
github.com/somu1795/baudrate

Yeah, I suspect that the dongle is a virtual com port, so it doesn't care about the baudrate as such. Does that chipset use the USB interface on the Ublox chip itself or a usb to serial converter as well?

I think the traditional way is to send a command at each baudrate until you get a response or a character echo. I am playing with this stuff on Arduino, so my approach will be to initiate commands from a cold start that try 9600, 38400, and then 115200, and immediately change the baud rate to 115200 before changing the other parameters.

I have measure the voltage on the 'battery' again, and its down to 2.3v on mine. So, its probably a cap.

Select to expand quote

boardsurfr said..

decrepit said..
Yep, I changed the rate to 5s, can really tell that that is what is happening, but disconnect and plug back in and its 1s.

Sounds like the battery is dead, and the module does not have flash memory. Not all ublox 8 modules have flash. Your module is supposed to be a NEO M8N, which would have flash. But some fake modules don't, seems like your module is one of them. It may still have an original ublox 8 chip inside, but it's apparently not a genuine NEO M8N. Not a big surprise, since ublox sells the NEO M8N at about $12 (for quantities of 250+), and that does not include the board the chip is mounted on.

You may want to see if you can replace the battery with a new one. If the chip "forgets" everything every time you unplug it, that has some other consequences, too. For example, you can't have a "hot start", which is faster. More annoying is that the reported time will probably jump by a few seconds when the chip finally gets the correct number of leap seconds, which can take 10-15 minutes.

These should have flash memory. I did a firmware upgrade on mine and it worked, and I assume its the same flash.
The warm start is much faster than a cold start, so I agree, the battery is really needed.

Select to expand quote

decrepit said..

boardsurfr said..

Sounds like the battery may be empty (if you're talking about the NEO-8 module). IIRC, the "real" NEO-M8N modules are supposed to have Flash memory to which they can save settings without needing a battery.

Can you enable GLONASS and GALILEO at the same time as GPS?

Yep, I changed the rate to 5s, can really tell that that is what is happening, but disconnect and plug back in and its 1s.

This model has a battery connected directly to pin 22, labeled "V_BCKP"




I guess you can't tell by the label, but there's the battery top left.

And yes it has all sorts of formats I've never heard of, including GPS and Glonass, not sure about Galileo, but I think so.

Did you buy your GPS from the link Daffy provided? I have two from that supplier, and there are a few differences. On the boards I have, if the battery/capacitor is soldered with the tab the way it is in your photo, it cannot work. The two tabs to the south of the picture are the ground, and the top of the battery is the positive. When I look at the module in better light I can see that the two tabs are actually one pad, and the tab to the north is meant for the battery to be soldered to. Unless your battery is upside down?

Also, both my modules are marked as NEO-M8N-0-10 and yours is marked as NEO-M8N-0-01. The barcode is also different in layout. Something is different.

Also, the battery is probably a capacitor. Each time I measure it, the voltage is dropping over the day. Its down to about 1.9v from about 2.5v just after it was used. The module which I haven't used yet, but soon will, measures as about 0.1mv. So, it will be a capacitor, or a really really bad battery!

Other than that the board layout looks the same.

I went to Jaycar today and they are now selling a M6 GPS module for arduinos for $49. Its a worry when Jaycar carries everything. The quality is not always there, but if you approach it as if the modules they sell have no instructions and you need to do your own research, its not too bad.

I just checked my photos I took of the drone m8n module I bought. It has the same number and barcode style as the other 2 modules I have, again, different to yours.

I stumbled across this when looking at the GPS information:

rtklibexplorer.wordpress.com/2016/11/04/pi-zero-based-gps-data-logger/

This looks like a Rpi Zero using the serial interface direct to the GPS module instead of using USB and a converter. It looks like a better approach if all you have is serial GPS modules.

Checked my battery/capacitor at 0.2mv between the one nearest the chip and the one nearest the terminals. after it's been sitting all night.

checking between ground and pin 22 on the chip I get 0.1mv.

The ublox data sheet says an m6 pin 22 is normally tied to ground and when converting from a 6 to an 8, this is OK to stay that way.
So connecting anything there indicates some thing different going on.

When plugged in I get 3.25v disconnect drops to virtual 0.
So I'd better investigate the connections a bit further.

Can't remember where this came from, I think it's Steven's or maybe from a sellers site?





It has the same chip as mine, but the battery/cap is then other way around. If it's not a polarised cap that should be OK.

But the cct shows it as a battery.

Guess I'll just have to disconnect it and see if I can figure it out.

well here it is.



Looks like a 3.3v battery to me and thanks formula for spotting this, soldered in the wrong way around, which looks to have shorted it out.

The multi meter also says it's not wired the same as the schematic shows. The big pad closest to the chip is ground, but the connection closest to the mounting hole doesn't go directly to pin 22, there's a 1k resistor in between, so the battery is on the other side of the resistor

So the question is now, can I find another battery?
More later.
My google search found a rechargeable button battery, hadn't thought of that but it makes good sense. Probably doesn't do much harm to short out a rechargeable, so I've soldered it back in the correct way round, and wonders of wonders, it's slowly charging up. So I'll let it sit for a while and see if it works any better.

well looks like I fixed that and now nothing works @##@$%%^^^.

Am I getting some sort of message here?
now the Pi isn't seeing the dongle or the module.

This ancient brain isn't a help either, took me ages to wake up to the fact, I have to modify /etc/rc.local every time I swap them over.

And then much later, that you have to reboot for this to have any effect!
So now the Pi is seeing the module but only makes an 80byte file then shuts down. I can no longer get the module to output any data, with any baud rate with both receiver and port set the same, and it says I have "0" channels even though the green light is flashing. I can see my adjustments going in and getting acknowledged, so there is communication just no data.
So have I fried something with all my probing? I wasn't wearing my antistatic wrist strap, but I'm standing on slate tiles on concrete, in bear feet. Or have I set something wrong no that I can actually save something?

And The dongle seems to be working fine in u-center but isn't starting a file in the Pi.

Think I'll concentrate on something else for a while, until the smaller 2200mah battery and an otter box arrive.

Select to expand quote

decrepit said..
well here it is.



Looks like a 3.3v battery to me and thanks formula for spotting this, soldered in the wrong way around, which looks to have shorted it out.

The multi meter also says it's not wired the same as the schematic shows. The big pad closest to the chip is ground, but the connection closest to the mounting hole doesn't go directly to pin 22, there's a 1k resistor in between, so the battery is on the other side of the resistor

So the question is now, can I find another battery?
More later.
My google search found a rechargeable button battery, hadn't thought of that but it makes good sense. Probably doesn't do much harm to short out a rechargeable, so I've soldered it back in the correct way round, and wonders of wonders, it's slowly charging up. So I'll let it sit for a while and see if it works any better.

I hate to disagree, but I really think it is a capacitor. The markings suggest 3.3v 0.07 Farads.

I think the V_Bckp line needs to see VCC as well. There was something in the hardware manual about this.

Is it still not working? I wouldn't be surprised if that capacitor still works because the short shouldn't have affected it. (After re-reading your post, it sounds like you have resoldered it back in the correct way.)

I still haven't plugged in my second GPS module, and the voltage on that is still sitting below 1mV. If it charges up, it is a capacitor.

I will report back.

I am assuming the circuit is similar to this:

Select to expand quote

FormulaNova said..
I hate to disagree, but I really think it is a capacitor. The markings suggest 3.3v 0.07 Farads.

I think the V_Bckp line needs to see VCC as well. There was something in the hardware manual about this.

I know things have changed a lot in the 20 years since I was in the industry but it's a bloody small cap to be .07 farads, in my day it was micro farads.

And as I said, the ublox blurb says it's OK to be tied to ground.

But you may be right, the cct I have shows a battery like this.





But the way it's wired agrees with your cct not mine.
Any way whatever it is has taken a while to charge to 3V and now I can save settings, but not output data.

My second GPS module had the capacitor go up to 2.75 volts when running and then down to 2.5 once powered down. From a 'never-used' 0.1mv, I think we can assume its a cap.

The 0.07 farads is why they are called 'super' capacitors.

Its not a bad choice in this application as it will keep the ram powered up for long enough, but after a while it loses its config and satellite info.

That's bad luck to get a module that has had the cap installed the wrong way.

Now that you have removed the cap, make sure you start from the default baudrate of 9600.

Good luck.