meteohub.de

Since the meteobridge software latest update releases for vers. 3.7, APRS/CWOP has not accepted my data. From the Oct 1 release comes this statement: "slightly increases longitude and latitude precision when using CWOP weather network or other APRS services. We removed a not so appropriate rounding in the formula used for coordinate transformation." This is the error message I now receive from APRS/CWOP: "This report was not parsed as a valid APRS message: Invalid uncompressed location."

In trial and error, I found that the loran conversion of the decimal latitude of my location that was entered into the system tab of the meteobridge software (I have the weatherbridge unit) is adding a third decimal place to that conversion and uploading that to APRS/CWOP. This is happening in the latitude only. My latitude is 38.06664, but this then gets converted and sent to APRS/CWOP as 3803.100N. CWOP/APRS, if I am correct, only accepts two decimal places, and thus rejects the data packet. If I change my latitude in the meteobridge interface to 38.067, the conversion goes through to APRS/CWOP as 3804.02N, and the data packet is accepted. Longitude of -120.64788 converts to 12038.87W, and is accepted. This has taken me a few days of lost uploading to discover what was going on. It seems that there must be a way to have the loran conversion of the five-decimal-place latitude conform to APRS/CWOP requirements.

Is this just my problem, or is there a bug in the update as to the rounding formula for coordinate transformation?

The term LORAN as used here is a misdemeanour as the coordinates have absolutely nothing to do with anything LORAN as the lat/Long are simply dDDMM.mm and it really should not matter how many decimal minutes follow the whole minutes as if a system requires 2 then it should simply ignore the rest but APRS/CWOP don't appear to do this and to some extent something they don't handle well.

However in regard the MB and your coordinate examples 38.06664 without rounding should be 3803.99840, rounded to 3 decimals 3803.998, rounded to 2 decimals 3804.00 which is very different to 3803.100 which appears to be as a result of rounding the decimal minutes only and not accounting for the whole minutes.

38.067 -> 3804.02 straight up with only zero's after the .02 (38.04.0200 etc) so rounding has no affect to 2 decimal minutes

-120.64788 -> 12038.8728, rounded to 3 decimals 12038.873, rounded to 2 places 12038.87 and there no actual need to "mathematically" round anything as it's simply being rounded down, dropping the insignificant values.

If the MB is coming up with 3803.100N then Boris will need to look at this as it should be 3804.00N

Thanks for the reply. Sorry for my ignorance, and for the "misdemeanor." Not sure how my mistaken understanding is at the level of a crime, but okay. I'll be sure to post less until I've become an expert in all this. However, your calculations are the same I get. And it's not "If the MB is coming up with 3803.100N," since that is in fact what is being reported. So my ignorance aside, I'm grateful someone will be looking into the issue.

JackOtter, The misdemeanour comment was certainly not towards you, it was toward the APRS and CWOP use of a rather incorrect term so don't take that as any offence towards yourself, or ignorance on your part, that was not the intent of my comments.