[Paco, Yuta]

RF FPMI is recovered after c1sus DAC-0 card replacement

Summary:
 - We wanted to check if FPMI locks after DAC-0 card relacement (40m/17620).
 - 60 Hz noise similar to what we saw in February prevented us from locking FPMI stably, but fixed it by turning off FM9 of coil output filters in MC1 and MC3 (40m/17462).
 - There are slight changes in locking gains, but it now locks reliably.

FPMI locking:
 - MICH: 1 for REFL55_Q, MICH_GAIN=18 (used to be 11) gives UGF of 45 Hz
 - DARM: 1 for AS55_Q, DARM_GAIN=0.044 (used to be 0.04) gives UGF of 134 Hz
 - CARM: 0.567 (used to be 0.496) for REFL55_I, CARM_GAIN=0.011 gives UGF of 224 Hz
 - Attachment #1 shows all the OLTFs.

60 Hz noise:
 - FPMI locking was not stable, and we moved back to YARM locking to see if 60 Hz noise is higher or not.
 - Attachment #2 shows 60 Hz noise measured with MC_F and YARM. The noise was actually similar to what we saw in 40m/17461, so we checked MC1 and MC3 dewhitening
 - FM9 of coil output filters was turned on for some reason (probably because of burts we were doing when fixing c1sus). MC1 and MC3 FM9 ELP28 filters should be off.
 - This made FPMI locking stable and 60 Hz noise lower by more than an order of magnitude (Attachment #3).

Last week I noticed that the high power amplifiers in the Frequency Generation Box became hot after 2 hours of continuous operation with the lid of the box closed. When I measured their temperature it was 57C, and it was still slowly increasing (~< 1K/hr).
According to the data sheet, their maximum recommended temperature is 65C. Above that their performances are not guaranteed anymore.

These amplifiers aren't properly dissipating the heat they produce since they sit on a plastic surface (Teflon), and also because their wing heat dissipator can't do much when the box is closed. I had to come up with some way to take out their heat.
The solution that I used for the voltage regulators (installing them on the back panel, guaranteeing thermal conduction but electrical isolation at the same time) wouldn't be applicable to the amplifiers.

I discussed the problem with Steve and Koji and we thought of building a heat sink that would put the amplifier in direct contact with the metal walls of the box.
After that, on Friday I've got Mike of the machine shop next door to make me this kind of L-shaped copper heat sink:



On Saturday, I completely removed the wing heat dissipator, and I only installed the copper heat sink on top of the amplifier. I used thermal paste at the interface.

I turned on the power, left the lid open and monitored the temperature again. After 2 hours the temperature of the amplifier had stabilized at 47C.

Today I added the wing dissipator too, and monitored again the temperature with the lid open. then, after a few hours, I closed the the box.
I tracked the temperature of the amplifier using the temperature sensors that I installed in the box and which I have attached to the heat sink.
I connected the box temperature output to C1:IOO-MC_DRUM1. With the calibration of the channel (32250 Counts/Volt), and Caryn's calibration of the temperature sensor (~110F/Volt - see LIGO DOC # T0900287-00-R), the trend that I measured was this:



Conclusion
The heat sink is avoiding the amplifier to overheat. The temperature is now compatible with that of the other component in the box (i.e., crystal oscilaltors, frequency multiplier).
Even with the lid closed the temperature is not too high.

Two things remain untested yet:
1) effect of adding a MICA interface sheet between the heat sink and the wall of the chassis. (necessary for gorund isolation)
2) effect of having all 3 amplifiers on at the same time

I am considering opening air circulation "gills" on the side and bottom of the chassis.

Also we might leave the box open and who ever wants can re- engineer the heat sink.

For posterity.
- Ideally we would like that the heat sink had the largest section area. A brick of metal on top the amplifier would be more effective. Although it would have added several pounds to the weight of the box.
- We need these amplifiers in order to have the capability to change the modulation depth up to 0.2, at least. The Mini-Circuit ZHL-2X-S are the only one available off-the-shelf, with a sufficiently low noise figure, and sufficiently high output power.

There were several parts in this box which did not have shunting capacitors across their input power lines.  Only the four RF amps (ZHL-2) had them.

I soldered two capacitors (100 microF electrolytic and 150pF dipped mica) across the power supply lines of each of the following units:  11MHz oscillator, 29.5 MHz oscillator,  Wenzel 5x frequency multiplier and the 12x RF amplifier (ZHL-1HAD).

It was quite difficult to reach the power inputs of these units as some of them were very close to the inner walls of the box.  To access them I undid the front panel and found that there were several very taut RF cables which prevented me from moving the front panel even a little.

I had to undo some of the RF cables and swap them around till I found a solution in which all of them had some slack.  At the end I checked to make sure that the wiring is in accordance with the schematic present here.

This is how the RF generation box might soon look like:

A dedicated wiki page shows the state of the work:

http://lhocds.ligo-wa.caltech.edu:8000/40m/Upgrade_09/RF_System/frequency_generation_box#preview

The RF multiplexer is configured as shown in the figure. It is now effectively a 15x1 RF mux.

To select a required channel:

Run the script as shown below 

/opt/rtcds/caltech/c1/scripts/general/rfMux.py

>python rfMux.py ch11

For channel 10 to 16, you can just enter the required channel number and it is routed to the output.

For channel 1 to 8, you only need to input the required channel number as above. No need to run the code again to select ch9 after selecting ch1-8

How the NI-8100 controller works:

Whenever any channel of one switch is selected, the output of the other switch is set to its ch0 (ch1 and ch9 in the figure).

So selecting ch1-8 will automatically select ch9 as output for the other switch. IF you send a command to select ch9 afterwards, the first switch will be automatically set to ch1 and not stay on what you had selected before.

This post pertains to the fiber-coupled diode laser mounted in rack 1Y1.

To turn the laser on, first turn the power supply's key (red) to the clockwise. Then make sure that the laser is in "current" mode by checking that the LED next to "I" in the "Laser Mode" box in lit up. If the light is not on, press the button to the right of the "I" light until it is. Now press the output button (green). This is like removing the safety for the laser. Then turn the dial (blue) until you have your desired current. Presently, the current limit is set to around 92 mA.

To turn the laser off, dial the current back down to 0mA and turn the key (red) counterclockwise.

All of the LSC RF PDs have been aligned.  I didn't really change much of anything, since for all of them, the beam was already pretty close to center.  But they all got the treatment of attaching a Voltmeter to the DC out, and adjusting the steering mirror in both pitch and yaw, finding where you fall off the PD in each direction, and then leave the optic in the middle of the two 'edges'.

Before aligning each set (PO, Refl, AS), I followed the procedure in Rob's new RF photodiode Wiki Page. 

Also, for superstitious reasons, and in case I actually bumped them, I squished all of the ribbon cable connectors into the PDs, just in case.

 In the order that makes more sense to me, it looks like you have:

REFL11, REFL33, REFL55, REFL165,

AS55

POX11

POP22

We don't really need POP22 right now, although we do want the facility to do both POP22 and POP110 for when we (eventually) put in a better PD there.  Also, we want cabling for POP55, so that we can illuminate it after we re-install it.  If we're working on 2f PDs, we might as well consider AS110 also, although I don't know that there was a fiber layed for it.  The big one that you're missing is POY11.

 A new RF cable has been included for POY11. Cabling for POP55 and POP110 might or might not exist. I will check and report it.

[Koji, Jenne, Kevin]

Jenne worked on fixing REFL11 last week (see elog 4034) and was able to measure an electrical transfer function. Today, I tried to measure an optical transfer function but REFL11 is still not responding to any optical input. I tried shining both the laser and a flashlight on the PD but could not get any DC voltage.

I also completed the characterizations of POX. I redid the optical transfer function and shot noise measurements. I also took a time series of the RF output from the PD when it was powered on with no light. This measurement shows oscillations at about 225 MHz. I also measured the spectrum with no light which also shows the oscillations at 225 MHz and smaller oscillations at ~455 MHz.

The plots can be found at http://lhocds.ligo-wa.caltech.edu:8000/40m/Electronics/POX?action=show.

This is looking better, but the fit data for the TF should be plotted along with the data. The data should be made up of points and the fit a line.

For the fit, we should have the Q of the main resonance as well as the peak height of the main resonance and the values of the gain at the notch frequencies.

Also the peak as well as the notches should have the frequencies fit for and labeled. In principle, you can make the plot on the wiki have all of the data. Then in the end we can print the plot in a small size and glue it to the PD's backside.

 The measured power levels of the RF source harmonics are given below:

We are considering inclusion of bandpass filters centered on 11 and 55 MHz  to suppress the harmonics and meet the requirements specified in Alberto's thesis (page 88).

As seen in the previous measurement the first harmonic of both the 11 MHz and 55 MHz outputs are about 30dB
higher than desired.  In an attempt to attenuate these and higher harmonics I introduced SBP-10.7 filters into
the 11MHz outputs and SLP-50 filters into the 55 MHz outputs.
Then I measured the height of the harmonics again and found that they were suppressed as expected.  Now harmonic
at 22 MHz is 58dB lower than the 11 MHz fundamental.  And the 110 MHz is lower by 55 dB compared to the 55 MHz
fundamental.  None of the higher harmonics are seen => they are below 70dB

SLP-50 has an insertion loss(IL) of 4.65 dB and Return Loss(RL) of 3dB.  It would be better to use SBP-60
(IL=1.4 dB and RL=23dB)

The filter on the 11 MHz lines is okay. The SBP-10.7 has IL=0.6 dB and RL=23 dB.

The SLP-50 filters which were on the 55 MHz lines have been replaced with the SBP-60.  Their respective characteristics are given below:

SBP-60 has lower insertion loss and higher return loss.  

This may however change the phase of I and Q in the demod boards and they will therefore need to be readjusted.  Currently the output power level of 55 MHz demod is at 2dBm, whereas it ought to be at 6dBm.   I have not yet corrected that.  Once that is completed Kiwamu will adjust the phases.

 I shifted the temperature sensor to a new location.  See the photograph below.  I noticed that the higher temperature is reached on the side where there are two RF Amps.  So it would be better to check the temperature of that  area and make sure that it remains well below 65 deg.  The operating maxium is 65deg C

Here is a picture of the new RF source layout.

And here is a photograph of it

RF Source box has been mounted in the 1X2 rack. 

Heliax cables have been directly attached to the box and anchored on the side of the 1X2 rack.  Here is a list of Helix cables which have been connected so far.

RF Amp operating temperature

Earlier measurement reported by Alberto in LIGO-T10004-61-v1 based on the LM34 temperature sensor were lower than that shown by placing a calibrated thermocouple sensor directly on the heat sink by about 5deg C. The difference probably arose because the LM34 was located on a separate free-hanging copper sheet attached to the RF Amp by a single screw, resulting in a gradient across the copper strip.   I tried to move the LM34 which was glued down, but broke the leads in the process.  I then replaced it with another one mounted much closer to the heat sink and held it down with a copper-strip clamp.  There is no glue involved and there is heatsink compound between the flat surface of the LM34 and the heatsink.  Picture attached. 

  The picture also shows the new filters which have been put in place to reduce the harmonics.  Note that the SBP-10.7 which was to go on the 11 MHz Demod output is located much farther upsteam due to space constraints.

I have posted the attached RF status update and 1Y2 rack layout to the svn.

For the photodetector frequency response project, our new RF Switch Chassis (NI pxie-1071) arrived today. I took the switches out of the old chassis (Note for future generations: you have to yank pretty darn hard) and put them in the new chassis, which I mounted in rack 1Y1 as pictured. 

The point of this new chassis is that its controller is compatible with our control room computer setup. We will be able to switch the chassis using TCP/IP or telnet, aiding in our automation of the measurement of photodetector frequency response.

We need a distribution unit in the LSC rack to: 1) collect the demod signals coming from the Frequency Generation Box 2) adjust the power level 3) generate 2nd harmonics (for POP) 4) distribute the demod signals to the single demodulation boards.

The base line plan is the following:

The box can be build up gradually, but the priority goes to these parts:

I need help for this work. I know exactly how to do it, I just don't have the time to do it all by myself.

 Besides the Distribution Box, the demodulation part of the upgrade would still require two steps:

1) upgrade the Band Pass Filters of the demodulation boards (I have all the parts)

2) cabling from the distribution box to the demod board (one-afternoon kind of job)

I have prepared several diagrams outlining the current state of the RF System.

These are uploaded into the svn40m here  and will be kept uptodate as we complete various parts of the task.  These plans have taken into account

the new priorities of the LSC (set out by Koji here )

We (Koji, Kiwamu and I) took stock of the RF cables which we have inherited from the earlier RF system and have made new plans for them.

I took stock of the filters purchased for the modifying the demod boards.  We have pretty much everything we need so I will start modifying the boards right away.   The following table summarises the modifications

We seem to have a spare SHP-175.  I was wondering where that is supposed to go. 

This is the status and tentative schedule for completing the various tasks.  I have put the dates based on priority and state of the hardware.

The RF Cable layout plans are drawn on top of a Lab Layout.  The various subsystems are drawn (not to scale) on separate layers.  The graffle files are located here  .  I thought they might come in handy for others as well.

I have made several changes to Craig's script for better pythonism. Its more robust with different libraries and syntaxes and makes a tarball by default (w/o a command line flag). These kinds of general util scripts will be going into a general use folder in the git.ligo.org/40m/ team area so that it can be used throughout the LSC.

I don't think we need/want a coherence calculation, so I have not included it. Usually, we use coherence to estimate the uncertainty, and here we are just plotting it directly from the dist of the sweeps so coherence seems superfluous.

I very badly forgot to log about this in the crush of surfs.

I removed Koji's proto-beatbox RF comparator amp from the X arm ALS setup.  I was investigating hacking it onto one of the discriminator channels in the new beatbox, now that Yuta/Koji's Yuta/Koji's phase tracker is making the coarse beatbox path obsolete.  Upon further reflection we decided to just go ahead and stuff the beatbox board for the X arm, and use the proto-beatbox to test some faster ECL comparators.  This will be done first thing in the morning.

In the meantime the old amp is in my cymac mess on the far left of the electronics bench.

I pulled out the RF amplifier box from the IOO rack and swapped the amplifiers for FOL beat frequency amplification. The earlier gain of 62dB (ZFL500LN + ZFL500LN) was reduced to 40dB gain (ZFL500LN+ZFL2AD).

I also swapped one of the broken sma cables that was connecting the two amplifiers with a good one. Front ports of the module were relabeled and the box was put back on the rack.

During the course of this work, I had to turn OFF the green BBPDs on the PSL table to protect them and they have been powered up after putting the module back.

Today I was around the IOO rack.

I shutdown the power to the beat PDs on the PSL table and disconnected the D sub 3w3 connector that was powering the RF amplifier panel on the IOO rack.

I moved the RF amplifier from the panel to a box that can go on the rack. The box will also hold the RF amplifiers that will be used for FOL. I have not completed putting in all the amplifiers. But the RF amplifier for ALS is in place and the box has been installed on the IOO rack for locking tonight. The power supply to the green beat PDs has been switched ON.

I took the out of loop noise measurements for ALS X and Y and the attachment is the screenshot of it (X and Y have rms of ~300Hz and ~400Hz).

I had to touch the Y end steering mirrors for green to get maximum GTRY before making thes measurments.

I pulled out the RF amplifier box from the IOO rack again and added the new RF amplifiers for FOL. 

I replaced one of the decoupling capacitors of the ALS beat note RF amplifier ; the poloproylene capacitor with a ceramic capacitor (0.1uF) .

After putting back the box, I confirmed that we had a beat note. I did not get a chance to measure the ALS noise after putting back the box because the IFO was already occupied.

I will post a detailed elog of the components in the RF amplifier box once I am done with it (hopefully tomorrow)!

The components of the RF amplifier box are in place. The RF amplifier box has been mounted on the IOO rack and the front panel connections have been labeled. Attached is the photo of how things look in the inside for future reference.

Sometime in the next few days the box will be pulled out to replace the panel mount SMA barrels in the front with insulated ones.

As Koji found one of the spare channels of the ALS/FOL RF amplifier box nonfunctional yesterday, I pulled it out to fix it. I found that one of the sma cables did not conduct.

It was replaced with a new cable from Koji. Also, I rearranged the ports to be consistent across the box, and re-labeled with the gains I observed. 

It has been reinstalled, and the Y frequency counter that is using one of the channels shows a steady beat freq.

I cannot test the amplitude of the green X beat at this time, as Koji is on the PSL table with the PSL shutter closed, and is using the control room spectrum analyzer. However, the dataviewer trace for the fine_phase_out_Hz looks like free swinging cavity motion, so its probably ok.

The RF analyzer was returned to the control room. There are two beat notes from X/Y confirmed.

I locked the arms and aligned them with ASS.

When the end greens are locked at TEM00, X/Y beat amplitudes were ~33dBm and ~17dBm. respectively.
I don't judge if they are OK or not, as I don't recall the nominal values.

 I gutted one of the $2 red laser pointers to build a laser source whose amplitude we could modulate at RF frequencies. Basically, I cut off the bulk of the housing from the pointer and soldered a BNC connection into the two terminals that the 2x 1.5V batteries were connected to. When I applied 3V across this BNC connector the diode still worked. So far so good.

Next I added a bias tee to the input. I put 3V across the DC input of the bias tee and added a -3dBm signal into the RF port of the tee. The laser beam was incident on a PDA100A (bandwidth of 1.7MHz) and, sure enough, Kiwamu and I could see a flat response in the amplitude at a given drive frequency out to around 1.7MHz.

We should check the response on a faster PD to see how fast the laser diode is, but we should be able to use this now to check the RF response of the green beat note PD. 

TO DO:

1. Add some capacitors across the DC input of the bias tee.

2. Do something about the switch on the laser diode.

3. Attach some labels to the laser that specify what is the required DC voltage and the maximum acceptable RF modulation amplitude.

I spent part of the afternoon cleaning up the area next to the Mode Cleaner where we keep all of our RF stuff:  Attenuators, BNC/SMA/LEMO adapters, Mini-Circuits items, and all sorts of other things which are useful while looking at our electronics/RF stuff.

We got another set of "Lyon" drawers, which aided in the organization process....Bob ordered 2, so we now have a 'spare' drawer set if anyone can think of something else to organize (unless this was premeditated for optics or something else?).

As you can see in the picture, (1) it's no longer a total disaster over there, and (2) some of the drawers have sub-divisions to make it faster and easier to find what you're looking for.  Please help out by putting things away in their proper place, and adding more labels or dividers to the drawers if there's something else which needs a 'spot'.

Awesome.

I propose that anyone who tries to do this kind of thoroughgoing cleaning should make an e-mail to call everyone available to join just for some hours
because every member has a responsibility to keep the lab organized.

And we have the list of things to do: Electronics (now it is halfway) / Cables / Optics / Screws / Tools ...

 [ Nichin, Eric G]

RF cables have been installed between deomodulator output PD RF MON and the RF switch for the following PDs:

 REFL33, AS55, REFL55,REFL165,REFL11,POX11,POP22

The cables are labelled on both ends and have been run on the overhead tray.

The cabling looks neat on 1Y2, but not so much in 1Y1(RF switch). I will better organize them later.

There were quite a few more demodulator units labelled with PD names. Do any of them need to be included in the automated frequency response measurement system? Please let me know so that I can include them to the RF switch and check them for proper illumination, which i will do for all the above PDs next week.

Test run:

I tested the RF switch selection code and then the data acquisition code for the NWAG4395A network analyzer and they both seemed to work fine. I selected the channel to which AS55 is hooked up to and then remotely got its transfer function.

There is quite some noise in the system as the plot shows. Especially the phase. Maybe my driving power was a bit too low. Have to figure out the reason behind this.

Further work:

• Make sure all the PDs are properly illuminated.
• Create a DC voltage reading's database for all PDs.
• Canonical plots for each PD to compare with the current data.
• Implement a script to fit the transfer function and extract required information about the PD.

The RF cables have been routed incorrectly. The cables run to the module from the front of the rack. We cannot close the doors to the racks if they are to remain this way.

I have asked Nichin to reroute the cables properly.

 [Nichin, Eric G]

As mentioned in Elog 10062, we found RF cables running between demodulators in rack 1Y2 and RF switch in 1Y1 to have bad SMA connectors (No shield / bad soldering / no caps).

we pulled out all the cables belonging to PD frequency response measurement system , 8 in total, and all of them about 5.5m in length.

Their labels read :

REFL33, REFL11, REFL55, AS55, POX11, REFL165, POP22 and POP110. 

All of them are now sitting inside a plastic box in the contorl room.

On another note, instead of fixing all the cables ourselves, Steve and Eric G decided to order custom made RF cables from Pasternack as professionally soldered cables are worth it. We have placed an order for 2 cables (RG405-550CM) to check out  and test them before we order all of the cables.

RF cables have been rerouted from the side of the rack, under the supervising eye of Manasa.

I moved the red ladder from near 1X4 to 1Y1 and back again.

Current list of RF cables:

REFL11, REFL33, REFL55, REFL165,

AS55

POX11

POP22

POY11

I have not connected them to the RF switch yet. ( until I figure out how to get both the switches working properly)

This Red Ladder movement also probably included some bumping of the MC2 stack (be more careful when working in the lab). Around 5 PM today, the MC lost lock.

When I came in later, I found that the MC was flashing the TEM17,9 mode and had been misaligned like this for ~1 hour. I looked through the MC SUS sensor trends and moved MC2 back to its old position to get the locking back. I disabled the WFS, tweaked the TEM00 mode alignment using only MC2 sliders, and then re-enabled WFS. Its been OK for the past 5 hours.

 I went into the lab and connected the RF cables to the Mux. Will take measurements for each PD henceforth.

 The input impedance of the resonant box was measured when an RF combiner was attached to the box.

Indeed the combiner makes the impedance more 50 Ohm and reduces the reflection.

**** measurement conditions ****

* The output of box, where the EOM will be connected,  was open so that the box tries resonating with a parasitic capacitor instead of the real EOM.

* ZFSC-3-13, a 3-way combiner from mini circuit, was used.

* The S-port of the combiner was directly attached to the box with a short connector (~ 30 mm).

* Port 1 and 2 are terminated by 50 Ohm.

* The input impedance was measured on port 3 with AG4395A net work analyzer.

* Reflection coefficient 'Gamma' were calculated from the measured impedance 'Z' by using an equation Gamma = (50-Z)/(50+Z).

The resonances are found at 11, 29 and 73 MHz (55 MHz resonance was shifted to 73 MHz because of no EOM).

Note that the resonances are at frequencies where the notches appear in the reflection coefficient plot.

Don't be confused by a peak at 70 MHz in the impedance. This is an extra resonance due to a leakage inductance from the transformer in the circuit.

An RF combiner should be included in the triple resonant box because it eases impedance mismatching and hence lowers undesired RF reflections.

Therefore we should use three cables to send the RF signals to the box and then combine them in the box.

(RF combiner)

 With proper terminations an RF combiner shows 50 Ohm input impedance.

But it still shows nearly 50 Ohm input impedance even if the source port is not properly terminated (i.e. non 50 Ohm termination).

This means any bad impedance mismatching on the source port can be somewhat brought close to 50 Ohm by a combiner.

  The amount of deviation from 50 Ohm in the input impedance depends on the circuit configuration of  the combiner as well as the termination impedance.

For example a resistive 3-way splitter shows 40 Ohm when the source port is shorten and the other ports are terminated with 50 Ohm.

Also it shows 62.5 Ohm when the source port is open and the other ports are terminated with 50 Ohm.

In this way an RF combiner eases  impedance mismatching on the source port.

(RF signal transfer at the 40m)

 According to the prototype test of the resonant box it will most likely have a non-50 Ohm input impedance at each modulation freqeucy.

If we install the resonant box apart from the combiner it will create RF reflections due to the mismatch (Case 1 in the diagram below)

The reflection creates standing waves which may excite higher harmonics and in the worst case it damages the RF sources.

 To reduce such a reflection one thing we can do is to install the combiner as a part of the resonant box (Case 2).

It will reduce the amount of the mismatching in the input impedance of the resonant circuit and results less reflections.

A rule we should remember is that a cable always needs to be impedance matched.

I realized that my impedance matching theory on an RF combiner was wrong !

In fact an RF combiner behaves more like an attenuator according to a reflection measurement that I did today.

A 3-way combiner reduces power of an input signal by a factor of 4.8 dB because it can be also considered as a 3-way splitter.

So it is just a lossy component or in other words it is just an attenuator.

(reflection measurement)

To check my speculation that I posted on #4504 I measured reflection coefficients for both cases.

In the measurement I used a heliax cable, which goes from 1X2 rack to the PSL table with a length of about 10 m. Note that this is the cable that had been used as '33 MHz EOM'.

At the input of the heliax cable it was connected to a direction coupler to pick off reflections and the reflected signal was sampled in AG4395A.

The other end of the cable (output side of the cable) was basically connected to the resonant box.

Then I did a reflection measurement for both cases as drawn in this entry (see #4504).

  - case 1 -  the combiner was inserted at the input side of the heliax cable.

  - case 2 - the combiner was directly attached to the resonant box

On the combiner, ZFSC-3-13, the port 1 and 2 were terminated with 50 Ohm, therefore the port 3 was used as an input and the source port is the output.

Here is a resultant plot of the reflection measurements.

Note that whole data are calibrated so that it gives 0 dB when the output side of the heliax is open.

There are two things we can notice from this plot:

 (1) The reflection coefficient at the resonant frequencies (where notches appear) are the same for both cases.

 (2) Over the measured frequency range the reflections were attenuated by a factor of about 9.6 dB , which is twice as large as the insertion loss of the combiner.

These facts basically indicates that  the RF combiner behaves as a 4.8 dB attenuator.

Hence the location of the combiner doesn't change the situation in terms of RF reflections.