Topics

Practicing "Safe RF" RF: DC block, RF Attenuator and/or limiter?

Konrad Roeder -- WA4OSH
 

My primary interest in my new (used) Tek 492BP spectrum analyzer is to test ham radio transverters, linear amplifiers, filters and antennas with a return loss bridge. At my job, we're urged to always practice "Safe RF" to prevent damage to sensitive RF instruments. I'm not quite ready to buy a DC Block/Limiter that costs as much as what I spent on my Spec-An, but I still would rather blow a fuse than my beloved Tek 492.

Tell me more about what DC blocks, limiters and/or external attenuators you use on your instruments when making measurements.
--Konrad

Tom Gardner
 

If you spent less than $21 on a working Tek 492, you are lucky :)

The Minicircuits BLK-18-S is small enough to be fitted internally, and has a bandpass of 10-18000MHz. Internal fitting means it can't be forgotten and can't be knocked off, but different internal cables may change the overall frequency response.

A DC block isn't a panacea, but it is a useful prophylactic.

https://www.minicircuits.com/WebStore/dashboard.html?model=BLK-18-S%2B

On 20/10/19 05:40, Konrad Roeder -- WA4OSH wrote:
My primary interest in my new (used) Tek 492BP spectrum analyzer is to test ham radio transverters, linear amplifiers, filters and antennas with a return loss bridge. At my job, we're urged to always practice "Safe RF" to prevent damage to sensitive RF instruments. I'm not quite ready to buy a DC Block/Limiter that costs as much as what I spent on my Spec-An, but I still would rather blow a fuse than my beloved Tek 492.

Tell me more about what DC blocks, limiters and/or external attenuators you use on your instruments when making measurements.
--Konrad

Greg Muir
 

Konrad,

If you are not extremely familiar with using a spectrum analyzer for the applications that you mention, here are some suggestions.

1. Do not connect a transmitter or any high RF level directly to the input of the spectrum analyzer. I have seen people do this several times not knowing the proper procedures when using an analyzer for RF signal analysis. The result can be very costly in repairs to the instrument.

2. Either built-in or external overload protectors are nice but they do have their limitations mostly from a maximum RF power level. Some external devices may be slow enough so as to allow a significant power level to pass on to the test equipment before activating protection.

3. If the situation permits, use an antenna connected to the spectrum analyzer. This can be done if you are not performing precise testing of power levels and such or not overly concerned with harmonic and spurious levels.

4. If you must connect to the output of a device under test to the analyzer input, utilize an in-line power attenuator to reduce the RF to a level lower than the maximum allowed for the spectrum analyzer. (Example: https://birdrf.com/Products/Test%20and%20Measurement/Attenuators.aspx). Calculate the attenuation required in relation to the RF power level you will be measuring.

5. If access to the transmission line is possible, use an indirect method of looking at the RF such as a signal sampler or something similar. These devices connect in-line and have a capacitive probe (sometimes adjustable to set the sampled signal level) with a sample port to connect the analyzer to. To see what is offered Google “RF signal sampler.” But you must be careful because some sampled signal levels may exceed the analyzer safe RF input level so use of an in-line attenuator between the sample port and analyzer may be required.

I carry a field kit containing various attenuators with different attenuation and power levels. Much of my work involves dealing with large broadcast transmitters that have included RF signal sample ports on their transmission lines. But many of these ports will have several watts of power at the sample connector so it is standard practice to initially install attenuators in the line connecting the spectrum analyzer until the power level can be determined. Then any unneeded attenuators are removed to bring the signal up to the proper level required for testing.

For off-line testing I also utilize RF terminations (dummy loads) anywhere from 50 watts to 35 kW. In this instance I again use in-line signal samplers and various attenuators to match the sampled RF to the spectrum analyzer.

Since I start with an excessive amount of attenuation to begin with, I never worry about overloading the spectrum analyzer input nor any other device that I may be using in my tests. If you don’t know what RF level you are starting with it is much safer to start with a huge amount of attenuation then reducing it until you reach the proper level.

I seldom if never use the method of connecting any high RF power device directly to the analyzer input through attenuators.

Greg

Tom Gardner
 

On 20/10/19 08:35, Greg Muir via Groups.Io wrote:
I seldom if never use the method of connecting any high RF power device directly to the analyzer input through attenuators.
(Much sensible advice deleted for brevity)

I'd add that if you are using switched attenuators, be sure you understand the attenuator's characteristics when switching, as will be seen by not only the source but also the spectrum analyser.

Konrad Roeder -- WA4OSH
 

Greg,
To test amplifiers at work, I use a 50 Ohm attenuator with enough attenuation loss that the spectrum analyzer and power meter run near their "sweet spot". For most lab instruments that tends to be around 0-10dBm or 1-10mW. So to test a 50 Watt amplifier, I use a 40dB attenuator on the output side. For the input of the amplifier, I use a coupler to sample the input signal. I add the offset of the attenuator loss to my reading. The input power may already be at a low level like 0dBm, so the coupler might give me a signal at -30dBm for example. In the same way, I add the coupling factor loss to my reading.

But I still want to guard my spec-an against a DC bias should I have a leaky cap or a bias-tee connected for some odd reason. I also use a power limiter should my coupler or attenuator fail.

--Konrad

Konrad Roeder -- WA4OSH
 

Tom,
I agree. For my hobby, I don't have to have a brand-new keysight DC-Block and power limiter. Most likely, my choices of attenuators and couplers will be such that my spec-an is running in an acceptable power range.

The risk of having a failed cap or accidentally having a bias-tee connected can indeed ruin my afternoon. Using a DC block is probably highly warranted.

Although I'm initially sticking to 904, 1296, 2304 and 3456 MHz ham bands, I'm eventually interested in experimenting at 10GHz and above. A broadband DC block like the BLK-18-S will work for me as long as I'm down in the lower four microwave bands.

--Konrad

Konrad Roeder -- WA4OSH
 

Yes. Attenuators are not flat and normalization does not fix the actual levels. It's only trace math. Attenuators also come with the penalty of eating the precious dynamic range of my instrument.

Greg Muir
 

Konrad,

Thank you for your reply. I am now aware that your knowledge regarding the proper use of test equipment is very good so please accept my apologies if I assumed otherwise in my last reply.

As for DC blocks, I took a quick look at what is available on eBay and found that there is a wide array one can find for significantly low prices that appear to be of reasonable quality. One could always characterize the item before use so as to get an idea of the “true” frequency response, VSWR and so on.

A sidebar to Tom Gardner’s input regarding Mini Circuits products – I realize everyone desires to have the HP/Agilent/Keysight name on their test accessories but many times I have found that Mini Circuits product specifications often exceed those of HP/A/K including RF amplifier products especially in regards to noise figure. It is an excellent source for RF related products.

Excessive RF protection devices is another matter with respect to cost. Again you might look at the line Mini Circuits offers.

I’m not a promoter of Mini Circuits products although I have used their things for many years. I have become more impressed with their offerings as time goes on especially with the performance-to-cost aspect. When you can spend 10-20% of the cost of a HP/Agilent/Keysight product and get the same or better performance it certainly takes the sting out of purchasing an item.

Greg

Sergey Kubushyn
 

On Sun, 20 Oct 2019, Greg Muir via Groups.Io wrote:

I second that. Mini Circuits stuff is really good and performance / price
ratio is probably better than anyone's else. And they make lots of different
things.

Unless one needs something really exceptional Mini Circuits products would
probably satisfy almost any need.

Konrad,

Thank you for your reply. I am now aware that your knowledge regarding the proper use of test equipment is very good so please accept my apologies if I assumed otherwise in my last reply.

As for DC blocks, I took a quick look at what is available on eBay and found that there is a wide array one can find for significantly low prices that appear to be of reasonable quality. One could always characterize the item before use so as to get an idea of the “true” frequency response, VSWR and so on.

A sidebar to Tom Gardner’s input regarding Mini Circuits products – I realize everyone desires to have the HP/Agilent/Keysight name on their test accessories but many times I have found that Mini Circuits product specifications often exceed those of HP/A/K including RF amplifier products especially in regards to noise figure. It is an excellent source for RF related products.

Excessive RF protection devices is another matter with respect to cost. Again you might look at the line Mini Circuits offers.

I’m not a promoter of Mini Circuits products although I have used their things for many years. I have become more impressed with their offerings as time goes on especially with the performance-to-cost aspect. When you can spend 10-20% of the cost of a HP/Agilent/Keysight product and get the same or better performance it certainly takes the sting out of purchasing an item.

Greg
---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************

GerryR
 

Would the SMT Station Monitor from www.preciserf.com help with your needs?
Regards,
GerryR

----- Original Message -----
From: "Greg Muir via Groups.Io" <big_sky_explorer=yahoo.com@groups.io>
To: <TekScopes@groups.io>
Sent: Sunday, October 20, 2019 2:25 PM
Subject: Re: [TekScopes] Practicing "Safe RF" RF: DC block, RF Attenuator and/or limiter?


Konrad,

Thank you for your reply. I am now aware that your knowledge regarding the proper use of test equipment is very good so please accept my apologies if I assumed otherwise in my last reply.

As for DC blocks, I took a quick look at what is available on eBay and found that there is a wide array one can find for significantly low prices that appear to be of reasonable quality. One could always characterize the item before use so as to get an idea of the “true” frequency response, VSWR and so on.

A sidebar to Tom Gardner’s input regarding Mini Circuits products – I realize everyone desires to have the HP/Agilent/Keysight name on their test accessories but many times I have found that Mini Circuits product specifications often exceed those of HP/A/K including RF amplifier products especially in regards to noise figure. It is an excellent source for RF related products.

Excessive RF protection devices is another matter with respect to cost. Again you might look at the line Mini Circuits offers.

I’m not a promoter of Mini Circuits products although I have used their things for many years. I have become more impressed with their offerings as time goes on especially with the performance-to-cost aspect. When you can spend 10-20% of the cost of a HP/Agilent/Keysight product and get the same or better performance it certainly takes the sting out of purchasing an item.

Greg

Don Bitters
 

I discovered years ago that some of the RF microcircuits in the HP Agilent test gear were relabeled Mini Circuits and other manufacturers of microcircuits (11729C for one), they may have been selected or customized for the HP Agilent use. I successfully repaired an 11729C with a Mini Circuits device when the HP parts was NLA.
Don Bitters

Dave Seiter
 

Another advantage of using a DC block is saving the SA input connector from so many connection cycles.  On high end gear, I'd rather wear out a DC block than the actual input.
-Dave

On Sunday, October 20, 2019, 03:28:12 AM PDT, Konrad Roeder -- WA4OSH <@WA4OSH> wrote:

Greg,
To test amplifiers at work, I use a 50 Ohm attenuator with enough attenuation loss that the spectrum analyzer and power meter run near their "sweet spot".  For most lab instruments that tends to be around 0-10dBm or 1-10mW.  So to test a 50 Watt amplifier, I use a 40dB attenuator on the output side.  For the input of the amplifier, I use a coupler to sample the input signal.  I add the offset of the attenuator loss to my reading. The input power may already be at a low level like 0dBm, so the coupler might give me a signal at -30dBm for example.  In the same way, I add the coupling factor loss to my reading.

But I still want to guard my spec-an against a DC bias should I have a leaky cap or a bias-tee connected for some odd reason.  I also use a power limiter should my coupler or attenuator fail.

--Konrad

Tom Gardner
 

Agreed.

On my 492 I've mounted the DC block internally (and not worried about the effects of the different cable), and have a N-SMA adaptor on the external connector. The latter is small enough that the front cover still fits over it, albeit tightly.

On 21/10/19 07:33, Dave Seiter wrote:
Another advantage of using a DC block is saving the SA input connector from so many connection cycles.  On high end gear, I'd rather wear out a DC block than the actual input.
-Dave

Greg Muir
 

RF connector savers…

For general purpose work I use the RFN-7654 N(F) to N(M) connector savers on my equipment. They are made by Pan Pacific and are not the most exacting adapter but work well at most lower frequencies. Pan Pacific states that they adhere to a MIL standard but that is only for the gold plated center contact. There are no parameters stated with respect to frequency, etc. They do look good examining them on a network analyzer though and are available in the range of $5 to $8 each if you shop around on the net.

There are more manufacturers of this product (Pasternack, Fairview Microwave, Coaxicom, etc.) and their items do have excellent high frequency ratings (up to 18 GHz) but you had better be prepared for sticker shock. BNC connector savers are also a good idea given the commonality and usage rate of this type of connector and their propensity to wear out more with the use of softer materials in their composition.

Obviously if you are doing any precision work in the GHz range it is best to remove the connector saver and connect directly to the instrument port. They are just another bump in the line impedance characteristics.

Greg