You may know Harlan Howe from his twelve years as publisher and editor of Microwave Journal ®, or from his 34 years as a Microwave design engineer and engineering manager, or from his service as an IEEE fellow and past president of MTT-S.

Now, although semi-retired, Harlan is available to answer your questions about RF and Microwave engineering. If he doesn't have the answer, he will find an industry expert who does.

Click here to submit your question now.

[Editor's Note, 16 December, 2003 : If you submitted a question to Harlan in the last several days, or plan to submit one this week, please be patient as we observe the Holiday season. The next Microwave Flash is scheduled for the week beginning 5 January, 2004 and the next installment of Ask Harlan will be published at that time. Thank you.]

FROM: Michael Wu, Askey

Dear Harlan:

I have a tough question about RF measurement. I am working in the industrial field and serve as an circuit test engineer. My question is : Is there any method to measure RF signal on PCB board level without soldering SMA connector? In digital signal, we can use scope probe to measure signals on the trace. However, if we use any probe to measure the trace between two components, it will cause mismatch and power loss problem.

In Lab or school, we can solder the SMA connector to measure RF performance. But in the commercial products, all the circuits are layout done. How can we know what happened inside the inter-stage circuirs?

This problem has bother me for a long time, is there any solution for this?




Dear Michael,

One solution is to add a loose directional coupler to the design of the board at the critical interstage points. You can then make a measurement at that point without disturbing the main circuit.

FROM: Mysore Jayasimha, Bharti cellular Limited

Dear Mr. Howe

So glad to write to you. We began our Career in RF and Microwave around 10 years ago. We studied your book Stripline Circuit Design by artech and we could build some very good Amplifiers & Oscillators.

In all the Point - Point Microwave radios the IF freq used for the Modulators are either 70 Mhz or 140 Mhz. I would like to know why it is so.



Bharti Cellular


Dear Jayasimha,

Early radar receivers used 30 or 60 Mhz. I suspect that 70 and 140 Mhz were chosen initially to get more information bandwidth. Once a frequency is picked and companies start producing standard products, everyone jumps on the bandwagon to save money and it soon becomes and industry standard.

FROM: Cliff Bayer, UBS


I am an associate at UBS Investment Bank, focusing on the Industrial sector. I have a few questions in reference to Microwave Amplifiers.

1) Is it safe to say that this technology started as in the military and has recently grown to the commercial sector?
2) What are some future trends you see in this sector?
3) What are some of the top performing companies in the amplifier sector?
4) What are some of the top performing companies in the component sector?




Dear Cliff,

Virtually all microwave technology began in the military sector starting with the MIT Radiation Lab during WW II. Commercial products started to appear in the '60s with private links, Television Electronic News Gathering and the like. When the military market started to dry up in 1990, there was a major push to find new markets and telecommunications was the answer. The search for new markets still continues today as the telecomunications markets have slowed. The future trend, as I see it, will be for low cost consumer products like "Bluetooth".

I am sorry, but I won't answer your last questions, since I will offend those of our paying advertisers, who don't get mentioned.

FROM: Sofia Vatti, University of Patras

Dear Harlan,

I need to calculate the exact jitter of a dual-loop frequency synthesizer.All the papers I've read about jitter are very theoretical and demand knoweledge of the statistic theory. I need to understand what the term jitter means in practice.Are you aware of a more practical way to calculate the jitter of even a simple PLL (for instance, through its transfer function).Please suggest any documentation (papers or books) that could be useful in this matter.

Thanking you in advance, I wait whith much interest your reply.

Best regards,

Sofia Vatti


Dear Sofia,

It is hard to separate jitter from statistical analysis since it is a semi-random phenomenon based on the noise performance of the loop and the frequency offset range. There is a book: Jitter in Digital Transmission Systems , Trischitta & Varma, Artech House, 1989, ISBN # 0-89006-248-X, which may be more theoretical than you want. There is also a software package that will calculate Time Jitter: PLL:Linear Phase-Locked Loop Control System Analysis , Unruh, Artech House, 1991, ISBN # 0-89006-522-5. This software sells for $249, which may be more than you wish to spend. Sorry, I can't do any better.

FROM: Abbas Amiri, Broadcom Corp

Is it possible to perform two single port measurements at Port1 of a two port system, once the Port2 shorted and once left open, and compute all 4 scattering parameters?




Dear Abbas,

No, you cannot make a two port measurement with only one port as you have described, since the measurement must include a through-port condition into a matched load.

FROM: Nafah Ash-Shebani, Al-makori telecom  

Hi Mr.Harlan,

How we can determine the dimensions of microstrip low pass filter,i.e, the length l & width w of each element?


Dear Nafah,

Most stripline and microstrip low-pass filters are based on stepped impedance transmission line equivalents of L-C ladder networks. There are many types. A good discussion and design procedures are given in: Microstrip Filters for RF/Microwave Applications , Hong & Lancaster, Wiley, 2001.

FROM: Chen Xuesong, National University of Singapore

Dear Dr. Harlan,

I have a project on 200W C band high power amplifer. But I am not sure which kind of RF PCB is the best to handle high power? Could you recommend one for me?

Thanks a lot,



Dear Xuesong,

You want the lowest loss material that you can find. I suggest one of the Teflon/fiberglass or Teflon/ceramic substrates, which are available from several of the suppliers in our Online Buyers Guide.

FROM: Rabindra Kishore Mishra, Berhampur University

I have got a LG Microwave Ovan. Its magnetron is getting damaged frequently. In the last 12 months, I have replaced it 3 times. Where can be the problem? Is it in the power supply section of the Magnetron? Or can it be due to low voltage (which is a frequent phenomenon here in India)? Or can it be due to over use (I use it for 1 to 2 hours a day)?


Dear Rabindra,

One or two hours a day is a lot of usage for a consumer microwave oven. The most frequent cause of magnetron failure is high reflection from an inadequate load and low voltage for the tube. You may want to look for a commercial or industrial oven, with an isolator, for your application.

FROM: Duncan Simpson, Simpco. Maint

Can EMF's cause swelling in hands,legs and feet? If so can you explain how?

Are there any ways to counteract the effects and block future attacks?

I have contacted several goverment agencys and have been met with "we cannot help" and "no one that can help." But we know that that is a lie: but that's the way it goes.


Dear Duncan,

There is no documented evidence that Electromagnetic Fields can cause the symptoms that you describe unless the power levels are so high that heating occurs, which you would be able to feel as discomfort throughout your entire body. Normal levels from broadcast stations, cell phones, radar, etc. are too low to have any effect. While various quacks have written nonsense on this subject, there is no scientific evidence to support them.

FROM: Hamidreza Memarzadeh, GPco.

Dear Harlan Howe,

I'm studying communication engineering in M.S degree. In my about Microstrip Antenna project I have faced some problems. I can not solve Dyadic Green Function with Method of Moment. Could it be possible to help my to over come my problem? Would you accept my thanks for your time and assistance?

Best regards

Hamdireza Memarzadeh


Dear Hamidreza,

While I am not a mathematician, I don't think the method of moments is the best way to solve the Dyatic Green's function field equations. There is an extensive discussion of Green's functions as related to microstrip antennas in: Microstrip Antenna Design Handbook , Garg, Bhartia, Bahl and Ittipiboon, Artech House, 2001, ISBN# 0-89006-513-6.

FROM: Chengheok Kang Madam, DSO National Laboratories

I understand that currently there is great conflict on the term IF and there is doubt of how could analog in RF be converted to digital format in order to create zero IF operation. Appreciate if you could give some details on the design concept making use of zero IF in system or applications. Warmest regard!


Dear Chengheok,

There are several ways to achieve a zero IF receiver. You can use direct detection or mix down to baseband using an IQ mixer output. The advantages are maintainence of the full information bandwidth. The disadvantage is the effect of 1/f noise, which can degrade the system.

FROM: Shivanand Ballurgi, Reliance Infocomm

What are the primary things required to carry out the Microwave Survey between two places?

1. System to be installed...........7GHz/34mbps.

2. Hop distance is approx...........100 kms.


Dear Shivanand,

There are 20 or 30 factors to be considered in a link design. I suggest that you look at: Microwave Radio Transmission Design Guide , Trevor Manning, Artech House, 1999, ISBN # 1-58053-031-1.

FROM: Zhang Kai, UESTC

Mr Harlan:

I have a confusion in my heart! Recently, I did some work about the widedband balun, but it made me mad! Some article said the length of the tapper balun is decided by the lower frequency of the passband. But other said it is estimated by the center frequency of the passband! Would you like to tell me how i decide the length of the balun! I simulated the balun design by me, but dB(S(2,1)) and dB(S(3,1)) far away 1dB from 3dB,can you give me some detail on design the tapper balun!



Dear Zhang,

It depends on what you mean by "wideband". Octave bandwidth baluns should be 1/4 wavelength long at the center of the band. Multi-octave structures with long tapers should be 1/2 wavelength at the low end of the band.



I'm designing a band-pass filter at 2GHz using Miniaturized hairpin resonators. In that once we freeze the resonator geometry we need to determine the coupling coefficients from the filter specifications. The next step is to determine the relation between coupling coefficient and the seperation between resonators (in some cases the offset between resonators is also considered) for all the three cases of coupling (two electric couplings and one magnetic coupling). Actually I'm stuck in the last step. Can anyone guide me as to how I can create the nomograms of coupling coefficient Vs seperations using a commercial EM simulator, for all the coupling configurations.

Thanks and Regards



Dear Sanjeev,

MOst of the commercial EM symulators will do the job, however, once you know the coupling coefficients, you can calculate the dimensions from the proceedures given in: Microwave Filters, Impedance-matching Networks and Coupling Structures , Matthaei, Young and Jones, Artech House, 1980, ISBN # 0-89006-099-1.

FROM: Shuai Brian, Agit

Hi Harlan,

I'm a RF engineer , I am engaging in designing a mobile phone . Do you have some suggestions or experiences on designing RF shielding case ? Now I placed our shielding case onto the board , but unfornately when connected to the Base simulator , the phase error is very bad at some frequences , and very good at other frequences . Could you help me? Thank you.



Dear Brian,

It sounds to me like your circuit board is radiating and reflecting off the shield when you install it. As an experiment, try placing some lossy material between the board and the shield. While this is probably not a good long term solution, it should tell you if that is the problem.

FROM: Omid Darvishi, EERC


I’m a student of communication engineering. A question in radar has occupied my mind and I have no answer to it. I hope you can help me. I know that after pulse compression in radar receiver, the signal power is boosted by a factor of PC processing gain (Gp). However using pulse compression makes the receiver matched filter much wider that is Gp times as great as matched filter bandwidth in conventional radars, so more sensitive to noise, that is the noise power in the receiver becomes Gp times more. This means that no improvement in SNR can be achieved in comparison with conventional radar with equal transmitter energy. Therefore as a conclusion using pulse compression in radars gives no advantage against noise jamming ECM (when the jammer bandwidth is great enough) but I have seen in many texts that noise jamming is the least effective jamming technique against PC radars, because of PC processing gain, that is PC makes jamming to signal ratio less by a factor of Gp. Well where is the mistake in my deduction?

Thanks in advance,



Dear Omid,

The primary effect of pulse compression in a radar is to transmit a long coded pulse and to process the received echo to obtain a relatively narrow pulse. The effect is to increase to detection capability of a long pulse radar system while retaining the range resolution of a short pulse system. There is an extensive explanation and discussion of the parameters of pulse compression radar in Chapter 10 of: Radar Handbook , Skolnik, McGraw-Hill, 1990.

FROM: Aladin Assisi, Catron

Dear Harlan

My question is not a technical one, but I am asl\king you because I am sure that you have the answer.

Fifteen years ago, there was a famous company specialized in microwave antenna mwasurement. It was called "Scientific Atlanta". I know that it is no more existenet with this name. Can you please tell me where it has ended and how to contact the new company that takes its place and/or performs its activities?

Thank you very much.

Aladin Assisi


Dear Aladin,

The company that replaced Scientific Atlanta is MI Technologies, 4500 River Green Pkwy, Suite 200, Duluth, Georgia 30096-2580. Their web site is:

FROM: Bobi Catorski, Bobcomm

Dear Harlan

I have a power HF signal source whom I need converting into DC as efficient as possible.

the signal spec are :

signal source BW : 1~30 Mhz P:200W @50 ohm

can you advice me the best way to approach the solution.

Best regards



Dear Bobi,

As I understand your question, you want to convert a 200W HF signal to DC power. I'm not sure why you want to do this, since you have already lost efficiency in generating the HF signal. The only suggestion I can offer is to use a bridge of high power rectifier diodes. You can check with some of the semiconductor suppliers in our on-line buyers guide for availability of appropriate devices.

FROM: Liu Zuqi, Nanjing University of Science and Technology

How to design the transceiver of 3G mobile phone, for example, the system analyse, test, simulation and tool.

By the way, can you suggest me what kind of books should I read and what kind of websites I surf.

Thank you.

I am looking forward to hearing from your answer.


Dear Liu,

I can suggest the following texts:

Introduction to 3G Mobile Communications , Juha Korhonen, Artech House, 2001, ISBN# 1-58053-287-X,
and GSM, GPRS and EDGE Performance - Evolution Towards 3G/UMTS , Halonen et al., Wiley, 2002, ISBN# 0470-84457-4.

FROM: Alex Rachman, Dmatek

Dear Harlan,

Where I can find SMD antenna for 300-400Mhz?


Dear Alex,

We don't recommend specific sources in this column. However, there are a number of antenna suppliers in our Online Buyers Guide as well as the advertisements in Microwave Journal each month.