The affair in which connection is cut using CC85XXDK-HEADSET

Hi Fredrik,

Thank you for your reply.

Once again, may I ask a question?

I got the reply of the affair no longer linking when a substrate is brought close.

Although I think that evasion of saturation is possible by lowering the GAIN value of removing

CC2590 or transmission, I think that the distance which an electric wave reaches in that case will

become short.

I considered changing 7Pin (HGM) of CC2590 into Low from High as other ways.

Is a good result brought when this method is performed? Or is a bad result brought?

I can understand that RX becomes LowGAIN mode by setting HGM to Low.

However, there is no telling how setting it in LowGAIN mode influences an electric wave. Is LowGain Mohd used by what kind of case?

Please let me know.

Best regards,

Hirokuni oota

Hi Kjetil,

I'm sorry.

 I am mistaken in it and have written your name.

And I would like to ask a question one more.

Although the cause got a reply called saturation, can it measure and check somewhere?

For example, suppose that I look at some signal waveform of a chip.

Can it check being saturated from the signal waveform?

Thank you in advance for your help.

hirokuni oota

Hi Hirokuni-san, 

You can look at our PurePath Commander tool:
http://www.ti.com/tool/purepath-wl-cmd

This is a PC tool that allow you to do basic RF testing. If saturation occurs it should show up when doing Rx/Tx testing on the Rx side as strange or flat RSSI values (have not tested so not sure how this will behave during saturation). Remember to use modulated TX output when doing these tests to get a accurate apple-to-apple comparison with the field testing you have already done. 

That is the best I can think of to check this, although I'm quite confident that saturation is the root cause of your problems. 

Regards,
Kjetil 

Hi Kjetil,

Thank you for the reply.

 I consider the given reply.

Then, how is the reply about the HGM pin asked before that?

Now, the HGM pin of CC2590 of CC85 XXDK-HEADSET is being fixed by High.

If this is set to Low, what kind of influence will occur?

 When it is able to change, does anything change by CC85 XXDK-HEADSET?

I am sorry to have asked a question repeatedly.

Best regards.

 hirokuni oota

Hi Again. 

Sorry, missed that question the first time around. 
HGM can as you said be pulled low and the Receiver portion of the CC2590 will then be in low gain mode (from ~11dB in HGM to 0dB in LGM). This will lower the overall sensitivity and can help with the saturation issue you see, although you will need to test to see how much affect this will be. This will reduce the overall range, but not as much as removing the CC2590 altogether.

When setting this low, please remember to select CC2590 (LGM) in the radio panel of the PurePath Wireless configurator. This will correct all internal RSSI based thresholds in the protocol.  

Regards, 
Kjetil 

Hi Kjetil,

Can this be done on the fly? I mean, changing the signal strength and sensitivity assuming the CC2590 is installed and host mode is used?

Thank you,

Nir.

Hi Nir, 

Changing the HGM of the CC2590 from an external MCU is possible.
Detecting Saturation might be difficult, but the NWM_GET_STATUS will return average RSSI on recent packets received from Master and you can play with this to see if you can detect this. 
What can not be changed is the internal RSSI thresholds used in our adaptive frequency hopping algorithm. There are thresholds on what we call 'listen before talk') that will be off and such the behavior might now be optimal, but presumably better than saturation. 

That being said, is saturation really that big of on an issue? Testing with boards and antennas in close proximity might not be the same case as final products where there are encapsulations and other things coming into factor. 

Regards,
Kjetil 

Hi Kjetil,

Thank you. I assume i can also control the signal strength and sensitivity of the CC85xx device itself on the fly, right?

The scenario i am looking into is this:

A cordless phone with a display sits on the base cradle. A call comes in (base is connected to PSTN)  and i wish, using the data-gram, to display the caller id on the wireless unit's display. if the RF link will not work, i assume no data link will work as well. The base antenna and wireless antenna will be less then 10cm apart.

Nir.

Hi Nir, 

Actually no. Output power is fixed and there is no methods of controlling the gain the RX stage. 

First I would have checked if putting the CC2590 in low gain mode helps on one of both side of the link. If you can detect that the phone is in the cradle then this should be relatively straight forward. 

Another alternative is to let the MCU control a power switch to the CC2590 or use some logic to override the PAEN and EN pins to completely disable the CC2590 when in the cradle. In this case the CC2590 will serve as a strong RF blocker, but since distance is short communication over ~10cm should not be an issue (depending on casing etc). You can test this effect by using our development kit and select range extender none (PAEN=EN=0) while still using our CC85XX+CC2590 module or Headset boards. 

Regards, 
Kjetil 

Hi Kjetil,

This sounds great. Thank you!

Perfect. I would have tested this well prior to putting a product in the market. A glitch in the 'cradle detect', especially upon picking the phone up, would be devastating as link performance with the CC2590 as a blocker is useless over 30-100cm. 

Kjetil
 

Hi Kjetil,

Can you please send me a friend request on the E2E? I need to discuss some technical issues of our project.

Thank you,

Nir.