QUESTION

Question

I have decided to use M20057-1 as an active GPS antenna. I am using Quectel L96 GNSS module. There is an internal chip antenna on L96 module and I have tried the performance of L96 without M20057-1 before. I have achieved to get location data while the device at outdoor but I couldn't get the location data at indoor. That's why I have decided to use an active antenna to improve performance. I have just received the PCB with M20057 and I am not able to get the location data for both indoor and outdoor with M20057-1. So It is obvious that there is a problem with the design of the printed circuit board. But I don't know that the problem is matching circuit(does it affect the performance that much), layout or schematic.  So Could you please review my schematic and layout to check is there any problem? I know that it quite hard to detect the problem in this way but even your approaches might be pretty useful. I am going to share screenshots of schematics and layout. Waiting for your responses. Best Regards.

Answer 1

Quectel L96

• Noise figure: <1.5dBGain (antenna): >-2dBi
• Gain (embedded LNA):20dB (Typ.)
• Total gain: >18dBi(Typ.) => Max gain !!!!!!

M20057-1

• GainLNA LNA Insertion Power Gain 18.5dB => Maybe the gain is to high

M20057-1 is placed in the right corner. M20057-1 is a magnetic loop and has a need for ground plane left and right. You offer nearly no ground plane o the right side of the antenna. Based on that the signal will be poor. the bandwidth of M20057-1 is 65 MHz only. Have you tuned at M20057-1 on the 1572 MHz?

More about loop antennas in my IoT M2M Cookbook http://www.gsm-modem.de/M2M/m2m_iot_cookbook/

Comments

For bottom layer you are right ground plane is not sufficient. However It's connected to inner layer which is all ground. So I thought that was okay. I haven't tuned it to 1572 MHz yet. I will send the product to antenova for matching service. Thanks a lot for your comment

Answer 2

What made you use a capacitor as one of the tunning components? The data sheets shows an inductor and resistor. I suggest if you dont have access to a network analyser, is to change the cap to a resistor and see if that does anything.
Also is your gps module seeing any satellites at all or are they tracking then dropping them due to really low SNR.

If you really have bad SNR's i would look at your ground plane and also your noise levels on the board.

Also i notice you have hardly any stiching around the antenna. You really need to get the gnd planes to a very low impedance. Also i would keep the matching components on one side as using vias (especially just one) is not goodespecially on matching circuits at this freq.

Answer 3

@Naumann

According to datasheet the ground plane under the active antenna package should be removed; the advantage of a magnetic dipole antenna is that works well on metal; if Antenova asks for metal removal under antenna could be a sign that is not a magnetic dipole antenna. I would be very interested to know the reasons for which you mentioned that the antenna should have left&right ground only. Thank you

@Edhelsoldier

I assume your pcb has four layers; the internal layers are not shown, so make sure you have the cut-out under antenna. If you have a 2 layer pcb then it could be that the antenna needs some metal around (because of the tuning and not because of radiation - you lose 3db from gain without metal under). The best thing would be to measure, if possible the return loss to see if your antenna is tuned. A 0 ohm resistor in tuning circuit means a short, use only caps and ind in tuning circuit.

Good luck and pls keep us updated

Comments

Here you can see the ground layers for the bottom, top and inner layer all their polygons are connected to ground.

I follow the rules for cut-out area under the antenna. I am using PNP transistor as a high side switch for powering both GPS module and active antenna as you can see from the schematic and allowable current of this transistor is 100mA. I just realized that the L96 GPS module requires 150mA peak current. It might be the reason why I can not get good results. I got in touch with antenova they told me to place 0 ohm to C44 and 3.9nH to L3 without analyzing the PCB (Of course It is impossible to tune the antenna remotely) So I tried various number of inductor values to place L3 such as 3.2nH,6.8nH,12nH,27nH,47nH, and 100nH. I couldn't find any correlation between the values and performance. I got the best SNR values when I leave L3 open circuit and place 0 ohm to C44.