blecm_setTxPowerInADV not working in SDK 2.1.1

Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
Anonymous
Not applicable

Hi BCM support,

Currently I am tuning the advertisement Tx power by blecm_setTxPowerInADV()

originally I used

blecm_setTxPowerInADV(4);

to get the max TX power, I roughly measured the RSSI by iSO app, it shows -36~-40dBm

When I change to

blecm_setTxPowerInADV(0);

It still shows -36~-40

And then I try

blecm_setTxPowerInADV(-30);

The measurement is still the same.

I also found the .tx_power_level in BLE_PROFILE_CFG structure is "UINT8" type, it will have some problem if we set the negative number in the parameter. Is there something wrong with the TX power setting functions?

0 Likes
1 Solution
Anonymous
Not applicable

Hi boont​,

I think I have verified the function blecm_setTxPowerInADV() is work. This topic can be closed now. I will continue test with the advertising signal, and will create another topic to talk about the signal issue.

Thanks.

View solution in original post

0 Likes
10 Replies
BoonT_56
Employee
Employee
500 likes received 250 likes received 100 likes received

I believed the Tx range for Class 2 is that between -6dBm and 4dBm. Therefore any value set below the minimum will be "round up" to -6dBm.

Below is a good reading:

Power level granularity in blecm_setTxPowerInConnection() and blecm_setTxPowerInADV()

0 Likes
Anonymous
Not applicable

Hi Boont,

Thank you.

I've check the article last week, but the Power_Table shows it can go to -25 dBm

That's why I am asking now.

In your words, the Tx power just can be set to -6dBm for minimum value?

Like below?

blecm_setTxPowerInADV(-6);

0 Likes
JacobT_81
Employee
Employee
250 replies posted 100 replies posted 50 replies posted

More likely than not, this is a problem with your means of testing. An iOS app's RSSI readout is not a definitive means of testing the true tx power. Do you have a network analyzer (connect to J11 of Tag3) to accurately measure the tx power?


Jacob

0 Likes
Anonymous
Not applicable

Ok,

I will check the real Tx power with RF engineers.

Thanks.

0 Likes

Btw, I ran past the internal developers on this and the feedback I got is that that chip can go as low as -20dBm. But it will round up to the nearest 4dB.

Is there a particular reason why you are pursuing this?

henryubnt

0 Likes
Anonymous
Not applicable

In the test field we may have 2 or more host device scanning for the devices.

When first time bonding, we will put the device and the host as near as it can, and we do not want some other hosts to scan the device and "rob" the bonding and connection. That's why we need an lower Tx power advertisement like a "whisper".

Anonymous
Not applicable

Hi boont,

We have test that the power, seems the blecm_setTxPowerInADV() function works.

The measurement shown the power is about 0dBm when we set blecm_setTxPowerInADV(4), and the measurement shown -9dBm when we set blecm_setTxPowerInADV(-6).

Here are 2 more issue:

1. The adv period=100ms It hard to get the 3 pulses in our measurement screen, should we have to change some option in BCM20737 for improving this problem?

2. As I know, the advertising channels are placed at 2402MHz, 2426MHz and 2480MHz. Below is one of our measurement, Is seems the channel 38 (The middle one) is placed in 2.428GHz. Is there any problem about the RF? (Please note that the 3 pulses didn't exist together, we have wait about 10 seconds to get this, why?)

IMG_2473.JPG

0 Likes

In a single advertising event, the transmitter will transmit once at channel 37,38 and 39 consecutively. At first glance, I am not sure why you didn't get C38 as 2426MHz. Perhaps you can repeat the above with the Hello_Sensor app running on the Tag3 and see what you get.

forrest.chien

0 Likes
Anonymous
Not applicable

Hi boont​,

I think I have verified the function blecm_setTxPowerInADV() is work. This topic can be closed now. I will continue test with the advertising signal, and will create another topic to talk about the signal issue.

Thanks.

0 Likes

Thanks for updating the thread henryubnt

0 Likes