Issues with beta units identified and fixed

[WHT]

UPDATE......

Mike just called me and verified my PRE-RELEASE BETA units had problems.

These were from the original batch of radios.

The low signal levels were caused by a problem with the N connector soldered to the circuit board.

The dead radio was caused by a shorted capacitor.

ALL OF THE PROBLEMS HAVE BEEN RESOLVED IN THE PRODUCTION UNITS CURRENTLY AVAILABLE FROM DISTRIBUTORS.



Can you test your Bullets on the bench. I tested a dozen of mine over the weekend and half of them had a bad RF section that added an additional 10 to 14 dB loss.

I suspected for several months there was a problem with some the Bullet M5 radios, but it was only last week that I finally had the chance to lay them out on the bench and narrow it down. I haven't reported my results to Mike yet, so I may be the only one that knows of this.

Here's my testing setup:

1) I calibrated a Bullet M5 to a known working Bullet as the AP with a known working Nano as the CPE.

2) I used that Bullet M5 as an AP and tested a dozen of my Bullet M5 as CPEs.

3) Most of the Bullet M5 CPE radios showed around -49 dBm received signal level from the AP and the AP reported about -49 dBm received signal level from the CPE (which would be the transmitted output of the CPE)

4) A few radios showed only -56 to -63 dBm received signal level from the AP and the AP reported about -56 to -63 dBm received signal level from the CPE (which would be the transmitted output of the CPE)

While points 3) and 4) together may indicate a simultaneous failure of both the transmitter RF output *and* the receiver front end, but I suspect its more likely a connection from the radio module to the antenna connector.

Several of the older Bullet M5 radios has a loose N connector center pin that wasn't soldered to the circuit board. But those radios had ZERO TX output and ZERO RX levels, not a mere 10 to 14 dB loss.

[kijoma]

hi,

this is a tad disappointing, back when the bullet5 first came out and i dared to show the insides (pic got removed at the time) >

http://arunvalley.net/bullet/naked.png

close up

http://arunvalley.net/bullet/dryjoint.png

I mentioned to Mike then that the solder joints to the N connector were "balled" and not flowed.. As a design / production engineer with oodles of years of experience this was a serious statement and should of been heeded.

Flowing such a large "heatsink" onto a PCB is difficult and it looks like they neither managed to get the solder to flow onto the Jack pins or the PCB. this is a production process issue as no doubt these had to be soldered by hand, no reflow or IR process is going to cope with that lump of metal..

The centre pin in the above is flowed nicely and is what the two ground pins should look like.. more heat was needed to flow the solder, give the chinese bigger irons :) (metcals or similar).

I will pop open a couple of the bucket of M5 bullets we now have and check this problem no longer exists.. As we wouldn't want something that changes power/sensitivity with temperature (as a dry joint like this could).

Cheers

[sar]

I agree. Cold solder or bad joints can be a nightmare, especially on an microwave frequency connection.

I deal with soldering on a daily basis so I would know .....

[WHT]

Its hard to tell how much wetting occurred on the two ground leads of the connector, but all the other portions look acceptable.

Keep in mind lead-free solder joints will cool with a grainy or even sandy appearance, and the wetting is not near as pronounced as lead alloy soldered joints which give the appearance of balling up. The only way to tell is if the flow has a concave meniscus; if it has a concave meniscus, then it didn't wet the substrate.

[kijoma]

hi,

come round here with all your big fancy words!

I have popped open two of the M5 bullets we have here and they are soldered a lot better than the old B5 above.

now the forum is fixed (thanks Mike!) i will post my 1st soak test results from the M5 tests i did.

[WHT]

"concave" is only two syllables, therefore not polysyllabic..so there! Phftt....

[marekm]

Keep in mind lead-free solder joints will cool with a grainy or even sandy appearance, and the wetting is not near as pronounced as lead alloy soldered joints which give the appearance of balling up.

Just wondering - there is a RoHS exception for "lead in solders for network infrastructure equipment for switching, signalling, transmission as well as network management for telecommunication". Shouldn't these devices qualify, so the good old tin-lead solder is still OK?

I (and no doubt many others) want high reliability, especially in devices mounted high on a mast where they may be costly to replace (not to mention the costs of downtime). And, more reliable actually means more friendly to the environment (less junk produced that needs to be recycled).