Webinars and lessons related to wireless IoT

Don't steal PCB antennas! Just stop by at Embedded World in Nuremberg, Hall 5, Stand 5-284 and ask politely. Or send me an email to harald.naumann (at) lte-modem.com. We will find a cheap way for PCB antenna, a cheap way for antenna consulting or consulting for IoT development or radio approval. You buy a few hours or days with me and I share my 35 years of radio engineering experience with you or with your team. At the Emde, everyone saves thousands of Euros.

Table of content:

• akorIoT Moon antenna stolen!
• akorIoT moon antenna in perfection
• NB-IoT, LTE-M, LwM2M, antennas, and radio approval as a webinar topic
• Consulting on antenna design and radio approval
• Meet with me at Embedded World in Nuremberg

akorIoT Moon antenna stolen!

Some PCB developers on LinkedIn are quite cheeky. They steal an antenna design up to and then pretend it is their work. The "experienced" hardware developer copies the KIcad logo onto the PCB. He "forgot" the request to copy the antenna. Antenna by www.akoriot.com on the PCB would have been enough for me. Not only did he copy it - he copied the antenna badly. Let's discuss the faulty antenna design here now. I marked the fault with numbers in the picture.

1) Ground plane for the antenna:

This antenna is a capacitively coupled inverted F antenna and requires a ground plane approximately 80-100 mm long. The PCB is much shorter and will therefore not work properly. Why this so has already been documented in my study "Low-cost PCB antennas for the wireless IoT". The study can be found on the server of akorIoT here: https://www.akoriot.com/white-papers/

After downloading, you have to read and understand 80 pages DIN A4. Reading and understanding 80 pages and trying out what you have learned naturally takes a lot of time. It is quicker to copy a layout with no sense. The last chapter of the study even tells you how to copy well. You just have to read it yourself.

2) Short-circuit point of the F-antenna:

As we have already learned with the DC current, a lot of current flows during a short circuit. It is the same with RF currents. A lot of current flows near the short-circuit point of an F-antenna. This current must be spread over the ground layers of the PCB with many vias. The many vias in the original design were forgotten. This makes the antenna worse.

3) Bent radiator:

In the original PCB, the radiator had to be bent for Sub-GHz because the PCB was not wide enough. On the PCB it is not necessary to bend the radiator. Bending or even bending twice is always a compromise. Bending reduces the bandwidth of the antenna.

4) PI circuit in the matching network:

The PI circuit has one component turned in one direction and the other in a 180-degree direction. As a result, the HF current must flow a longer distance. This often makes the matching worse.

Only one PI circuit was planned. A T-circuit with three components or a T-circuit with termination and thus four components was not implemented in the design. Whether you need a PI, T or T with termination is not known in the first design. How to do it correctly is already described in my IoT M2M Cookbook from 2014. You will find more on this in a few months in my "Antenna Cookbook".

5) Stripline with 50 Ohm:

Since the stripline to the cellular antenna and GNSS antenna was simply pulled through, it is reasonable to assume that it was not calculated at all, this leads to a mismatch at the radio module and cannot be adjusted. It leads to losses and HF energy flowing back to the radio module. Since the matching network is already wrong and the antenna was copied without modification, the probability is very high that the cellular radio module produces harmonics. This is only noticed in the radio certification in the test laboratory.

6) SIM card connector above the short-circuit point:

Exactly at this point, very high electric and magnetic fields will be generated. In addition, the cellular antenna will radiate in the area of the SIM card holder. In the original PCB, the SIM card connector was isolated from the ground plane to avoid interference, and in the poor copy, the SIM connector was placed where most interference is expected. The probability of malfunction is very high in the copy.

Summary:

Live and let live. After the counter-reaction on LinkedIn, the antenna thief changed his post and removed the copied antenna in the new post. But that doesn't change the fact that he simply stole the layout and has no idea about RF technology. He probably doesn't own a VNA and RMS worth 18000 Euros and uses the free KICAD instead of other PCB CAD programs with more features in RF design. KICAD is not bad, but not optimal for designing a GNSS tracker. The sum of errors in the design is long. I have mentioned 6 errors in the design and see at least 2 more errors. The PCB might work as a prototype, but it will certainly never go into mass production.

If you want to save a few thousand Euros, you can just order a few hours to days from me. I will gladly share my knowledge and you are welcome to transfer the money saved to those in need.

akorIoT Moon antenna in perfection

The antenna in the pictures above is internally called "moon antenna" because it looks like a half-moon. It is a very complex antenna structure and is not easy to copy. We need 4 to 7 working days, depending on the number of bands and the size of the PCB. If you invest many days, you will achieve the result in the pictures from approximately 600 MHz to 3800 MHz. The red curve shows the parameters of the antenna without the matching network. The blue curve shows the antenna with 2 components in the matching network. No or few components in the matching network also means no or few losses in the components for antenna matching. Any change in board size or enclosure will immediately change the centre frequency and frequency bandwidth. Changes in the antenna structure can compensate for this.

NB-IoT, LTE-M, LwM2M, antennas, radio approval as a webinar topic

In 2020, I offered webinars for the first time. Due to the pandemic, it was no longer possible to offer seminars. We spread the 6 hours of training over 4 days and trained twice for 45 minutes per day. At the end of each session, there were 10 minutes of questions and answers. Most questions from the chat were answered immediately after the lesson. All questions were summarised and answered by the next morning and distributed to everyone. As we repeated the training two weeks later, all participants were able to find a date.

I now plan to repeat the topics from 2020 and offer them in German and English. This will be followed by the supplementary seminar that the participants requested back then. So 2x 6 hours spread over 8 days for 300 Euros per participant. Between the first block of 4 days there will be a break and then the second block of 4 days.

Please send non-binding pre-registrations to harald.naumann (at) lte-modem.com. The pre-registrations will then determine the time zone.

Consulting on antenna design and radio approval

Wrong antenna design is one of the main reasons why a device does not get a certification. I am happy to help you choose the right antenna. I am also happy to explain how to integrate a PCB antenna into the main PCB. But I also explain at the beginning why an antenna chosen by the customer is right or wrong. In extreme cases, the antenna has to be custom designed and sometimes even simulated. There is a solution for the wrong copy in the pictures above. But the solution is not to copy an existing structure. How to evaluate and copy simple structures is described in my study called "Low-cost PCB antennas for the wireless IoT. NeoCortec www.neocortec.com ordered this study from me because their customers often copy the 868 / 915 MHz antennas incorrectly. Enquiries are welcome at harald.naumann (at) lte-modem.com

Meet with me at Embedded World in Nuremberg

From 21 to 23 June you can meet me at Crout GmbH, hall 5, booth 5-284. Besides antennas and radio modules we will show an LwM2M server and the possibility of low-cost radio certifications. I am looking forward to your visit.

Imprint

• Harald Naumann, Ludwig-Kaufholz-Weg 4, 31535 Neustadt, Germany
• Contact: harald.naumann (at) lte-modem.com, Phone: +49-5032 801 9985, Mobile:+49-152-33877687