Preliminary analysis of the G problem of the electronic competition

RF-刘海石

Preliminary analysis of the G problem of the electronic competition [Copy link]

 

For reference only

I am moving house recently and all the instruments and equipment have been packed, so there is no actual testing and debugging. This plan is for reference only.

I made this transceiver when I was in school, so it’s been 10 years, but there should be no problem with the principle. If there is any problem, please correct me.

This year's G question is a little more difficult on the FM transceiver. It needs to transmit 2 voice signals at the same time. It's not difficult. The difficulty of this question lies in the production of the transceiver, and it checks the mastery of RF circuits and analog circuits. I guess this question is relatively less, because making a transceiver requires equipment and instruments, and solid basic skills, otherwise it is difficult to do it in these 4 days.

Schematic diagram of the competition

The question requires the simultaneous transmission of 2 -channel voice signals. All that has been done before is the transmission of 1 -channel voice signal. I personally think there is no difference. The transmitter mainly adds an adder to synthesize the 2 -channel voice signals into 1 , and then modulates them. At the receiving end, they are demodulated and passed through a low-pass filter and a high-pass filter, or 2 band-pass filters respectively. However, band-pass filters are not easy to implement, so the 2- channel voice signals can be separated.

First, make an adder to combine the two voice signals into one . I personally think that this circuit can be realized with an ordinary operational amplifier.

Here I randomly selected an op amp as an adder to combine 2 signals into 1

I use a synthesis of 1KHz and 3KHz, other frequencies are also acceptable, but the two frequencies should be as far apart as possible, so that it is easier to use the filter to separate the two signals during the subsequent demodulation.

Next, make an FM transmitter. The difficulty of this question is the transmitter and receiver. This is also a test of pure analog circuits and the basis of RF circuits. If you are working in RF, the first question you will be asked in an interview is whether you have made a transceiver. This is my experience in interviews with other friends, and they will definitely ask you this question. No more nonsense.

I think that the old method of FM transmitter is to use PLL plus varactor diode. It just happens that the varactor diode is mentioned in the component list. Because the question requires the carrier to be 48.5MHz , the relative error cannot be greater than 1/1000 . I personally think that the phase-locked loop is just right for such accuracy.

The figure above is the principle block diagram of PLL

The main ones are the phase detector and voltage-controlled oscillator ( VCO ), which are the most critical. I made a transceiver when I was in school before, but it was a long time ago. I used MC145152 and homemade VCO . The advantage of this chip is that it does not require the control of an external MCU . It has 2 rows of parallel ports and can adjust the values of the N divider and R divider directly through the short-circuit cap. It is very simple to use.

VCO Circuit

This circuit can definitely be implemented, but it is worth debugging yourself. This is the VCO circuit I made when I was in school.

Phase detector circuit

This design uses the MC145152 chip to implement the phase-locked loop

Loop Filter

Active loop filter using op amp

Another thing is the voice amplification part. You must add a potentiometer to adjust the gain, otherwise the maximum frequency deviation of 25KHz is difficult to adjust. I won't say more here.

The voice signal is adjusted to the varactor diode. The amplitude change of the voice signal changes the capacitance of the varactor diode, which also changes the frequency of the VCO oscillator, so that the oscillation frequency of the VCO changes with the amplitude of the voice signal, thus realizing frequency modulation.

The power output of PLL is above 0dbm and is transmitted through the antenna, so the wireless transmission2mThe distance should be more than enough.

Receiver

Receiver Block Diagram

This design adopts superheterodyne mode.

The local oscillator still uses PLL, no external modulation is required, it is just a phase-locked loop for down-conversion. It still uses MC145152.

Bandpass filter. Since the antenna is used for reception, a bandpass filter must be added.

Give you a filter model, as for the parameters, you need to debug it yourself

Then there is the low noise amplifier LNA . This is an old design, so the components used are very old, but there is no problem with the solution. You can change it according to the components you have.

LNA uses 3SK122, a dual-gate field effect tube with low noise and high gain.

Mixer

The mixer can be an active mixer or a passive mixer. For the sake of simplicity, an active mixer is still used here, and a dual-gate field effect transistor is still used.

After the mixer outputs 10.7MHz intermediate frequency, it enters MC3361 for demodulation. In this way, the voice signal is demodulated.

But in the case of one signal, the voice signal can be output directly. If it is two- way voice signal, a filter must be used to separate the two signals. The simplest one is a low -pass filter and a high -pass filter, which can filter out the two signals, and then it can be done after voice amplification.

This is my plan at the moment, and I welcome everyone to discuss, analyze and correct me.

This content is originally created by EEWORLD forum user RF-Liu Haishi. If you want to reprint or use it for commercial purposes, you must obtain the author's consent and indicate the source

viphotman

This is well written. Is there any software that can simulate this solution?

1455173290

This antenna receives and transmits. Is this using a mixed channel common frequency band or a homemade antenna?

RF-刘海石

viphotman posted on 2019-8-8 09:24 This is very well written. Is there any software that can simulate this solution?

It can be simulated! You need to simulate! There is no need to use real devices to do it! Many devices do not have models! You can look at the PDF and see the internal structure! For example, for frequency dividers and phase detectors, you can directly use ideal models to make them! Learn and apply!

RF-刘海石

1455173290 published on 2019-8-8 09:27 This antenna receives and transmits. Is this using a mixed channel common frequency band or a homemade antenna? &n ...

The antenna is easy! Just make a copper wire! Requires a distance of 2 meters! This is too easy! There is no need to waste time on the antenna

小新拉布

I have a question, how do you get the initial parameters of the bandpass filter?

抬头就是阳光

I want to ask how to design the low-pass filter after the adder. We are going to use an RC passive low-pass filter circuit, but this circuit has a strong attenuation. We have made first-order, second-order, and fourth-order filters. We can't find any formulas for third-order and above filter circuits on the Internet. Does the author have any recommended circuit diagrams? Our cutoff frequency is about 4K.

RF-刘海石

Xiao Xin Lab posted on 2019-8-8 11:07 I have a question, how do you usually get the initial parameters of the bandpass filter?

Software simulation.

RF-刘海石

Look up and you will see the sunshine. Posted on 2019-8-8 11:45 I would like to ask how to design the low-pass filter after the adder. We are going to use the RC passive low-pass filter circuit, but this circuit has a relatively high attenuation...

If the passive attenuation is large, then use an active RC filter.

gmchen

Your solution has no problem with the high frequency part, but the superposition of the first two voice signals is wrong. It is two voices (each voice is 300 Hz to 3400 Hz), not two sounds of different frequencies, so they cannot be directly superimposed. You can refer to the solution of stereo modulation. I hope I will not mislead you.

generalcircuits

It's not easy to finish it in 4 days. Come on!

RF-刘海石

gmchen posted on 2019-8-8 15:28 Your solution has no problem with the high frequency part, but the superposition of the first two voice signals is wrong. It is two-way voice (each voice is 300 Hz...

OK, I may have misunderstood here, thanks for correcting me. Thank you very much! ! ! !

EwrdZzz

I would like to ask, how can I make the receiving circuit at the back track the frequency deviation of the transmitting circuit at the front? Can I change the PLL according to the deviation of the IF frequency?

mrf245

The RF part of this problem is actually very simple. It can be solved with a VCO+PLL, and even amplification is not needed.

The difficulty lies in two-way audio transmission.

There are 2 methods:

1) Digital mode (such as DMR) format transmission and reception, open dual time slot transmission. There are open source solutions on the Internet (search for open source MMDVM for details). But if you don’t prepare in advance for 4 days and 3 nights, don’t even think about it.

2) Use FM stereo transmitter chip, such as BA1404 BH1415 1417, etc., to modulate the two-way voice into two stereo channels.

Note that the separation of BA1404 is not high enough (based on my previous experience), while the separation of BH1415 1417 is good. When using 1415 1417, only the stereo encoding part inside can be used, and the synthesized audio signal can be directly modulated to the external VCO+PLL of your own. The step frequency of the PLL loop should be 100K or above.

The receiver converts the frequency to 10.7MHZ and then outputs it to the FM stereo intermediate frequency demodulation chip. There are various models. You can find a lot of information on the Internet.

I have not thought of a solution for the receiver tracking operation after the frequency fluctuation. I wish you all good luck.

Pluto-

Does anyone have a specific solution for Problem H?

taotao137

mrf245 posted on 2019-8-8 18:17 The RF part of this question is actually very simple, a VCO+PLL can do it, even without amplification. The difficulty lies in the two-channel audio transmission. There are two methods: 1 ...

The receiver uses a low intermediate frequency (fc = 200k, BW = 200k). The intermediate frequency signal passing through the limiter directly enters the phase detector to control the LO. Synchronization can probably be achieved.

ufo_xiao

Really good analysis

gmchen

EwrdZzz published on 2019-8-8 17:49 I would like to ask, how can I make the receiving circuit at the back track the frequency deviation of the transmitting circuit at the front? Can I change the PLL according to the deviation of the intermediate frequency?

This method is feasible. In essence, it uses a phase-locked loop to track the intermediate frequency to ensure the stability of the intermediate frequency.

This post has been modified. The previous one is not as good as the method here.