The Oscilloscope Diaries: One Man’s Trash Is Another Woman’s Broken Toggle Step Attenuator

With a reputation for putting out strong signals, a friend of mine came across an irresistible way to get me to tone it down…a Pasternack 0-82.5dB toggle step 1W attenuator. An ebay purchase from a dude who spelled ham “amatuer”. I pounced at the opportunity.

So, while it was being trucked across the country, I spent some time on thinking how I wanted to test it out. I was pretty excited about it…I mean, look at all those switches begging to be thrown?!

One night I came home from work and a box was sitting atop my mobile workstation (ahem, tv tray). AA1F and I sat down to shoot the breeze and drink a lovely glass of herbal tea while recapping our days. And there was a lingering promise of ice cream… But, truth be told, I was thinking about those damn attenuators. First off, the seller had actually sent me four of them when I had been expecting only one (bonus!) And I only wanted to just quickly check that they actually did something. I mean, flipping switches is a heck of a lot of fun it its own right, but seeing some results?! Bliss!

AA1F made a strategic error and went out into the kitchen to put his tea cup in the sink. Seeing my opportunity, I raced up the stairs with a trailing “Be right back!” Look, don’t judge, I meant it… I was going to simply get the satisfaction of seeing the voltage decrease with each thrown switch on the oscilloscope, and that would be that. Back chatting with the OM in no time!

(There is a moral of the story here…don’t trust sellers who can’t spell or friends that want to quiet you down…)

So, I excitedly queued up the Koolertron signal generator with a 1MHz 1V sine wave, attached a BNC jumper to the input of the first attenuator, attached another BNC jumper from the output of the attenuator and into the oscilloscope, and….

Well, first, I could not actually find my signal on the oscilloscope. This was most definitely operator error. I am still finding may way around this piece of test equipment.

After a bit of futzing (someone who reads this blog will appreciate this little bit of inside humor), I managed to get this…

Now I *love* my oscilloscope, but, I don’t really know how to use it all that well. And its nearly as old as I am. So my first electronically insecure thoughts, in this order, were 1) I screwed this up, or 2) the oscilloscope is busted. “Be right back” drifted into blatant lie territory, and a promise of a little KM1NDY handmade ice cream sundae were relegated to the dashed hopes department of AA1F’s imagination. I could hear the quiet steps of his feet retreating to his own personal space in the house. Well, had I actually been listening, I am sure I would have noticed that he was no longer waiting for me.

For the next, I don’t know, half hour, I turned up and down everything I could think of. More voltage, higher frequency. Lower frequency. Increased volts/div. Increased sec/div. Flip the attenuator switches. Flip them back. I mean its only time and voltage? Why could I not see a damn sine wave on the oscilloscope? Then I blamed the signal generator which clearly was not working. I took a coax jumper directly from the oscilloscope to the Koolertron…and….

Uh oh…

My fabulous new-to-me attenuator with a retail value (that I paid but a fraction) of nearly $1000 (each!) did not work. I put in a second one. Same thing. I put in a third one. Same straight line across the oscilloscope. No signal was reaching the oscilloscope. I took a deep breath, pulled out my multimeter. I checked for continuity between the center pins of the BNC connectors. I mean, I may not know much about electronics, but it seems to me that the center pins of each end of the attenuator should be in continuity with one another…

Not great news. There was no continuity between the center pins. I toggled some switches trying to get electricity to pass from one side to the other with no luck. I was now pretty confident that the attenuators, at least the first 3 of them, were broken.

Perhaps a bit out of order for the story, but for anyone who does not know what these things are, let me spell it out. Attenuators will take a signal, like the sine wave above on the oscilloscope that is a graphical representation of voltage plotted against time, and decrease the amount of that signal. The number underneath each switch on the attenuator represents a percentage (expressed in decibels) that the amount of radiofrequency (RF) power entering the attenuator will be reduced by. For instance, by throwing the “10” (10dB) switch, only 10% of the input power will exit the attenuator. You can look up conversion charts to get a better idea of this concept. For every switch you flick upward, you can add together the total decibels. For instance, if you turn on every switch, there will be 82.5dB of power loss through the attenuator. Keep in mind, I am only looking at voltage (versus time) on the oscilloscope, not power, but the same concept still applies, namely flicking switches should decrease the voltage of the outputted signal compared to the original signal.

Time for the fourth attenuator. Not holding my breath, I make the connections with the signal generator and o-scope. I flip all of the switches off and glance over at the oscilloscope display, and almost choke when I see it! A sine wave that traveled end to end through the one attenuator that may actually be working! Note, I made it so the sine wave would fill up as much of the display as possible. Again, voltage is on the vertical axis while time is horizontal. The number of ups and downs per unit of time is the frequency of the signal. I’ll keep the settings on the oscilloscope the same for the rest of this test.

I flip up the 0.5dB switch…

The waveforms are now a tiny bit shorter, representing a decline in voltage as expected with attenuation! Good news so far! If you count the number of peaks of the waves, you’ll note there are ten. This is exactly the same as the original picture before I flipped the 0.5db switch. So the number of waves per unit of time, i.e., the frequency, has not changed. This is exactly as expected in a working attenuator.

I flip the next switch up too. Now there is 0.5 plus 1 dB, a total of 1.5 dB, of attenuation. Again the height of the waveforms decreases showing that the voltage is reduced, with no change in frequency.

And the next switch for 3.5 dB of attenuation…

Now a total of 6.5 dB of attenuation…

Are you getting the point? With each newly thrown switch, the voltage exiting the attenuator is decreasing. Here’s 12.5 dB attenuation…

And finally, turning on one more step of attenuation (with the 0.5, 1, 2, 3, 6, and now 10 dB switches now turned on for a total of 22.5 dB) brings us to the lowest amount of attenuation that can be discerned with these settings on the oscilloscope.

Throwing one more switch (the first 20 dB switch) causes the oscilloscope to have a flat line. Decreasing the volts/div setting on the oscilloscope would “zoom in” on the signal and the sine wave could once again be seen. But, for the time being, I think you get the idea of how this attenuator is supposed to work.

Now with my quite rudimentary understanding of attenuation, I thought that these passive devices must be made up of mostly resistors. And that each switch must control an attenuation circuit that is in series with the ones adjacent to it. I figured if I opened up the box, I could probably trace the path of these individual circuits to see where they lost continuity. I have no doubt there is other, more engineer-like ways to describe this, but I am certain by now no one thinks I am an engineer.

So I took out the 8 tiny machine screws that held the metal plate in place and here is what I saw:

And a little closer up…

What I found was that each switch corresponded to a cell (now labelled 1 through 4) that indeed had a circuit of resistors within it. Each cell was in continuity with the cell next to it via the plates that are joined by pink dots below. And each of the plates that are joined by the green dots are also in continuity. Furthermore the green dot plates are in continuity with the pink dot plates.

However, when you throw the switch, say for cell number 2, the green dot plates become isolated and are no longer in continuity with the adjacent green dot plates, nor with the adjacent pink dot plates. This appeared to be the normal working condition.

I opened up one of the broken attenuators and began to check its internal continuity. Quickly I found adjacent cells that seemed to be abnormally electrically isolated from one another. However, with a little bit of pressure or changes in positioning of the attenuator, I noticed that the continuity would return. I suspected some mechanical problem that would keep opening and shorting as the handling conditions varied. Sure enough, it turns out the set screws holding several of the switches in place were loose. I found a wrench and tightened them down. Ta-dah! Continuity returned to the center pins of 2 out of the 3 broken attenuators. And when I tested them with the oscilloscope, it turns out they now worked! I am still trying to figure what’s wrong with the remaining broken one.

Now I do not know if the seller knew if these attenuators worked or not. I have already mentioned that he misspelled “amateur”, leading me to believe he may not be a ham. However his advertisement mentioned fox-hunting, and indeed I plan to use these for radio direction finding . In general the items were packaged quite nicely. I would think that you would need either an oscilloscope or a vector network analyzer to really see that something was going wrong with these, so it is probably fair to presume the seller did not know they were broken. Perhaps having the foresight to check for conductivity between the center pins of the BNC connectors would have been helpful. But more likely than not, some radio club had a box of these things that someone picked up at a ham flea market, gave them a try for amateur radio direction finding, and quickly realized that they did not work too well. Or perhaps they actually thought they worked *too* good, because once they put them on their antennas, they could no longer hear the signal at all. And they figured, someone out there may want these things…just go ahead now and put them on ebay.

Well, one man’s trash is another man’s trash, as AA1F is fond of saying. But in this case, this batch of switches provided me with one of the first opportunities I have had to troubleshoot an electrical problem with my oscilloscope. And a thank you to the person who thought of me when he came across these items. I was so grateful to you, that you almost got three of them sent your way… Just teasing, if you want one (a working one), its yours! They are as cool as I had envisioned them to be.

KM1NDY