Hello

I seem to be getting 1 out of 10-15 cm9/r52 deaf on arrival

Is that the normal rate?

Perhaps I am handling them wrong or something?

Thanks for any comments

Equis, what do you mean by 'deaf on arrival' ?
(wireless card are not recognized by RouterOS ? wireless cards do not rx/tx with enough power ?)

One of the the most common problems,
make sure that pigtail is not touching any metal part of the wireless card (try without pigtail).

sergejs:

Did you just say that when you plug the u.fl into any of the radio cards it should not touch the metal part of the radio card?

Because I bet just about everyone of my units the little cable wraps around and touches the metal part of the radio. There just so fragile. If so I'll have to go back and make some type of strain relive to keep them from touching.

-Michael

wireless cards do not rx/tx with enough power

Yup, About 15-20db down both ways.

I've seen about the same numbers. We test all our CM9s now before they get added to stock.

-=Russ=-

I've seen about the same numbers. We test all our CM9s now before they get added to stock.

-=Russ=-

Do you have any tool in order to make testing
procedure more efficient & accurate ?

Care to share ?

High Regards,
Andrei

Ehm...

Not really so much a "tool" as "interns."

I have a particular antenna/pigtail combo that reliably provides a -50db signal from our office AP when on the bench oriented properly.

"Plug the CM9 into the routerboard. Attach the pigtail to antenna port A. Power the board on. When it's booted, winbox in. Go to Wireless, select the interface. Go to Scan. Look for something floating around -50. If it's between -55 & -45 on average, it's fine. If it's -70, that's not fine. Label the bad ones for RMA."

A few hours later, the shipment is tested.

-=Russ=-

Interesting discussion:

Generally, you should always have some sort of antenna connected, otherwise the radio "sees" an open-circuit, where it should have a 50 ohm impedance load. open circuit means power reflected back into the radio, which potentially cause damage (on very high power radio transmitters with many W output, it does: probably less of an issue on a CM9)

We put together some "dummy antennas" with a 51ohm resistor on an N connector, which terminates the radio into approx the right load, and still radiates (not with any decent beam pattern, but it does) and with two of them placed about 1-2 feet apart, you get enough free space loss to create a "good" RF link. It's pretty repeatable too.

In our experience there are definitely batch yield issues on CM9s, I'm surprised more people don't complain. There are plenty of alternatives, including R52.

Would be great to hear more from other people -

Regards

CableFree Solutions

We are now having severe problems with the new CM9-GP cards. Both TX- and RX-wise.

Our failure rate on these cards is very high now. Anybody else with similar problems?