UBlox MAX-M8W-0-10 Module Thoughts

[Gandalf_Sr]

I've just started working on the MAX-M8W-0-10 and have designed and built a small breakout PCB (see picture).  The PCB uses a small 11 mF superCap as the Vbackup with theoretical life of 30 minutes.  I hooked it up to a USB-Serial adapter and monitored its output using the free Ublox U-center application.

My first impressions are WOW! In maybe 10 seconds* from cold, I had a fix with 6 satellites and that was with the antenna pressed against the inside of the window of my basement workshop (so I can only see half the sky at best).  After a minute I had 14 to 16 satellites, some GPS, some GLONASS, 1 SBAS, and a WAAS; the max I've counted is 19 so far (I think).  I have made zero changes to the configuration, just taken the OOTB setup.
[Edit] * The above 10 second time was not real; the module was being powered from a USB-Serial adapter it was hooked up to; that adapter was plugged into a USB 3.0 hub which, it seems, gives power out (with no LEDs lit) even when my PC is powered off!

Some questions and some ideas if I may...
1. How good is this accuracy-wise? Is it really good (compared to a Navsync CW25 for instance) or am I just out of date?
2. How accurate is the positioning? I Have seen the PDOP and HDOP numbers in U-center which are 1.2 and 1.5 in my early tests but how do I relate them to accuracy?
3. The time pulse signal output is configurable from 1 Hz to 10 MHz, given the MAX-M8W-0-10 uses an internal TCXO, is this anywhere good enough to be used as a GPSDO reference signal for test equipment?
4. Is the WAAS coming over a satellite or what? [EDIT] WAAS (wide area augmentation system) used to come over terrestrial transmitters but now it (also?) comes over dedicated geosynchronous satellites so I assume that the WAAS channel that pops up on U-center is indicating reception from one of those WAAS satellites.

[iMo]

3. The time pulse signal output is configurable from 1 Hz to 10 MHz, given the MAX-M8W-0-10 uses an internal TCXO, is this anywhere good enough to be used as a GPSDO reference signal for test equipment?

The signal could be used directly, the jitter will be rather large, however. I would not use it for a precision measurement.
Also doublecheck the signal shape at 10MHz - the NEO modules make "clean" signal when the output frequency is set to 48MHz/N where N=4,5,6,7,..
Better to use the output signal for disciplining of a quality OCXO.

[Gandalf_Sr]

Looking at the output from this which I've had running for days now, it's clear that it's a sensitive and powerful receiver.  I have a question about the 1 PPS signal that's output (by default)...

The 1PPS signal is a 100 mS wide positive going pulse with a 1 Hz frequency (duh!)  The question is, how do I relate the rising (or falling) edge to a specific time given that the data coming over the UART (or I2C) lines is going to take time to get to my controller? My guess would be that the rising edge marks a perfect zero second point but say I only need a time update from it once per second, how would I sync that to the 1PPS?

[ZigmundRat]

The time pulse alignment can be set to different values (GPS or UTC for example) and is also subject to settings such as antenna delay. The time message which is output tells you the time at the next PPS. So if the message says 2:43:03 the next PPS leading edge will be aligned with that time. In other words, the message is in advance of the PPS. Look in the manual under time-pulse for the details - it’s described more fully there.

Cheers!

[aix]

The question is, how do I relate the rising (or falling) edge to a specific time given that the data coming over the UART (or I2C) lines is going to take time to get to my controller?

See section 19.2 of  https://www.u-blox.com/sites/default/files/products/documents/u-blox8-M8_ReceiverDescrProtSpec_%28UBX-13003221%29_Public.pdf (unless I am misunderstanding your question, in which case my apologies).

[Gandalf_Sr]

Thanks guys, that makes a lot of sense; get the time over the UART and then zero the seconds as the pulse comes along.  I'll read up on those links.

[thinkfat]

How can it make a fix from cold start in 10 seconds? It takes around 20 seconds best case to receive the ephemeris data, for each SV, without which no position can be calculated.

[Gandalf_Sr]

How can it make a fix from cold start in 10 seconds? It takes around 20 seconds best case to receive the ephemeris data, for each SV, without which no position can be calculated.

I don't know but it has barely powered up for a few seconds before there is Lat/Long & Time displayed in the top right window of U-center.  Maybe it keeps some log of where it is and what I'm doing isn't strictly classed as a 'cold' start?

[thinkfat]

How can it make a fix from cold start in 10 seconds? It takes around 20 seconds best case to receive the ephemeris data, for each SV, without which no position can be calculated.

I don't know but it has barely powered up for a few seconds before there is Lat/Long & Time displayed in the top right window of U-center.  Maybe it keeps some log of where it is and what I'm doing isn't strictly classed as a 'cold' start?

Is the 10 seconds TTFF with your breakout PCB and the buffer capacitor? Then I guess it's internally storing the almanac and ephemeris data sets from the previous session. That will be a "warm start", then.

[Gandalf_Sr]

Except that my calculations on how long the superCap would keep the backup voltage up was around 30 minutes and the starts I've referred to are after it's been off for hours.  As my tests go deeper, I've realized that the 1PPS signal takes a while to come on which is I think a reflection of the M8 knowing that it hasn't got a reliable position as the datasheet says the 1PPS won't appear unless there's a 'lock'.  I'll run some timing tests on how long it takes from cold to for the 1PPS to appear and report back.

[Gandalf_Sr]

OK, it must have been a glitch in the matrix, I powered up a cold one and it was over a minute before I got the 1PPS signal (on an overcast day with the GPS antenna inside my basement window).  I ran tests that showed, when I removed power, that my superCap design held up the backup voltage for over 20 minutes (until it sank to 1.4V) and restoring power while the superCap was above 1.4V was like instantaneous for the GPS info / fix / 1PPS signal to come back.  I think what was confusing me was that the 3.3V supply comes through my USB-serial adapter from my powered USB3.0 hub, even if the PC is off.

[EDIT1] I confirmed that, even though there are no LEDs lit, my USB 3.0 hub feeds power to the USB-Serial adapter plugged into it and that has a small 3.3V LDO regulator that powers the GPS module.

[EDIT2] I did a timed test today, it took 2:30 (mm:ss) to get 1PPS signals from the module from a cold start running from a powered antenna inside my basement window.