This is a great, mathematically rigorous, approach which I find easier than using Smith Charts ! You helped me impedance match independent RF amp, Freq Conv, and IF amp modules in an experimental vacuum tube communications superhet radio that I am building for fun, with home wound antenna, rf interstage, oscillator coils and IF transformers. I was able to check my designs after building them with TinySA spectrum analyzer, NanoNVA network analyzer, and a 2 channel DSO with a built-in SIGGEN ! Thank you ! 🥰!
Great lecture, great examples, and as a hobbyist this video answered many of my lingering questions. And this mathematically rigorous treatment is much better than practical smith chart since it also reveals bandwidth implications and quantitative aspect of the matching network.
Sir you're absolutely amazing ! I'm currently reviewing these concepts for passing my exam ! And I was really stuck but I thank God I've found you 😊KEEP DOING FOR US VIDEOS PLEASE 😄
I hope you pass your exam! I wish I can make more videos here but pandemic pushed our institution to make a RU-vid channel. All my videos will be uploaded there now.
@16.09... If I use absorption method, but assumed values for X and B (series reactance and shunt susceptance respectively) to have + or - value respectively, and by taking - values for X and B in this case (to have series capacitance and shunt inductance), we get the same result... So, this appears to me like the resonance method is somewhat of a redundant method.
Hello Carleston. A question. So, in order to work with low Q values and wide bandwidths , do I need to divide the T and Pi network into sub matching networks in series?
Hello. T and Pi network can only push you to higher Q values. If you want low Q values, you need to use cascaded L networks. They look like cascaded Pi/T networks but it is more appropriate to think that they are cascaded L networks and between each network, you are matching the current load to some virtual resistance. With the correct values of virtual resistances, you will obtain a wideband match.
Rvirtual is a value of a resistance you need to achieve a specific Q-factor. Since we are designing the network to have a higher Q-factor, 2 elements are not enough. You need three elements to achieve a higher Q-factor. The middle element can be split into two. In the case of a shunt element, it will become two parallel shunt elements. For a series element, it will be two series elements. In between the split lies a virtual resistance that you will need to compute based on the Q-factor you need. Once you get the virtual resistance, you can compute for the values of the split elements and combine them into one (parallel combination for the shunt, series combination for the series).
Stub matching networks are not covered by this video. The two stubs are mainly used when there is no space in your circuit board for long stub matching networks. The position of the stubs can be varied along the lines but affixing them near the load end and source end ensures the smallest area needed.
Thank you so much for your excellent video! I am new to the RF matching network design. This video helps me to understand how to do calculations regarding 3 typical matching networks! I found a minor issue (probably a typo of a number) at 21:36. The Rvirtual should be 4.42ohm rather than 4.46ohm, so the Xs1 should be 4.6*4.42ohm=20.33ohm. Please kindly correct me if you see I am wrong. Thanks!
I do apologize for the very late reply. I hope this will still help. The best type of matching network will largely depend on your specifications. Moreover, aside from the networks discussed in this video, there are still a number of matching network techniques that especially work at high frequencies (above 1 GHz). The goal here should be to find the network that satisfies your needed specifications while minimizing the cost of implementation (number of components used). That is the best matching network.
Hello. By increasing the number of elements, you decrease the Q-factor of the circuit. In a chain of L-matching networks, the largest Q-factor of one L-network is approximately the Q-factor of the whole circuit. Successive L-networks can be used to maintain a small Q-factor. Look at the slide where the formula for the Q-factor is and try to calculate and simulate it by yourself.
Kareem Daraghma the quality factor (Q) of one L-network is related to the bandwidth (B) and resonant frequency (f0) by the formula f0=BQ. If we let f0 be constant, then decreasing Q means we increase B. Now, the quality factor for one L-network element is dependent on the load resistance (RL) and the source resistance (RS) by the formula Q=sqrt(RL/RS-1) if RL>RS and Q=sqrt(RS/RL-1) if RS>RL. Now, in between two individual L-networks lies a virtual resistance whose value can be flexible such that you can have a desired Q factor that is less than the Q factor using only one L-matching network. For example, consider two L-networks cascaded with each other. The source resistance is 50 ohms and the load resistance is 100 ohms. If we use one L network to match it, then Q=1. Now, if we set a virtual resistance Rv=75 ohms in between the two L-networks, the Q factor of the matching network from RS to Rv is 0.7071 and Q from Rv to RL is 0.5774. The total Q factor of the circuit is approximately the larger Q factor which is 0.7071. This is less than the Q factor of the single matching network only which is 1. You can simulate this using a circuit simulator and observe the effects.
Sir please if you can do for us a favor and make us a video for impedance matching using transformers as split capacitor circuit before 19th of november which is the day of my exam in this course ! I'LL BE SOOOOO GRATEFUL 💙💙
That is the first time I have heard of that. Also, all educational videos I will be making from now on will be uploaded to our Institute's official RU-vid channel. Apologies.
WTF! Is this a reading class because all this cat does is read. I wonder if he watches or listens to his videos because this one is got the worse humming with terrible background noise. My advice is that you invest in a decent microphone, they're not that expensive and stop using whatever crappy microphone you've installed in your system because it makes the video suck. I'm just saying ...
@16.09... If I use absorption method, but assumed values for X and B (series reactance and shunt susceptance respectively) to have + or - value respectively, and by taking - values for X and B in this case (to have series capacitance and shunt inductance), we get the same result... So, this appears to me like the resonance method is somewhat of a redundant method.
What you said about the absorption method becomes apparent when you use the Smith Chart for matching circuits. However, the resonance method exists as an alternative if you want to have a bandpass matching network that is not provided by the absorption method. Moreover, there is the issue of bandwidth where one method may have the better bandwidth specification compared to the other.
