Connecting multiple digital scanners to the same scanner antenna.

Status
Not open for further replies.

prcguy

Member
Joined
Jun 30, 2006
Messages
15,224
Location
So Cal - Richardson, TX - Tewksbury, MA
I think your post was for Ubbe, but I'll comment on this also. Any loss from SWR is reciprocal and will affect transmit and receive. The loss from bad SWR at the antenna or due to mismatch between 50/75 ohm cables, etc, is usually small and it may go unnoticed on receive but its still there.

Also, Stridsberg makes both active and passive splitters, so I'm guessing your comment on Stridsberg being the only answer would be for the active type, which I agree with unless you can live with a bunch of loss for the passive version.

Despite what you have shown here I must instantly disagree with what you have written. For receiving purposes SWR does not play a factor in regards to a scanner or any receiver. IF it is used for transmitting, say a ham radio, then I agree. You can attach a paper clip to the back of a scanner and be able to receive, however if done to a amateur radio and one attempts to transmit, then fat chance. I should know as I have had my ham radio license since 1968 and I seriously doubt things have changed in regards to standing wave ratio and reception issues. For accessing multiple radios a splitter really wouldn't work as a Stridsberg would be the ONLY answer.
 

prcguy

Member
Joined
Jun 30, 2006
Messages
15,224
Location
So Cal - Richardson, TX - Tewksbury, MA
I'm a little disappointed nobody caught the 9dB loss for a 4-way divider here which is wrong, it should be around 6dB. I'm not feeling well and that's my excuse but I remeasured the setup and found with three ports unterminated this divider has several dB of ripple across the passband and at 146Mhz it just happened to have 3dB more loss that normal. I terminated all unused ports and this 4-way divider now measures a fraction of a dB over 6dB using it as a splitter or combiner.

This again proves that using a splitter backwards as a combiner will incur the same loss as using it as a combiner, and that you should always terminate any unused ports on your splitters!


Ok, I'm home sick today but just for Jon I grabbed a 4-way power divider, signal source and power meter to show how a splitter works backwards as a combiner. The power divider is an Anzac DS-409-4 and perfect for scanner use with a 10MHz to 2GHz frequency response and good RF specs. The signal source is around 0dBm at 146MHz, well within the useful range of the power divider.

Pic 1 shows the output level of just the 146MHz signal source at .33dBm.

View attachment 76738

Pic 2 shows the splitter connected as a splitter with the signal source outputting .33dBm into the common port and -9.07dBm is left at one of the splitter outputs.

View attachment 76739

Pic 3 shows the splitter connected as a combiner with the signal source outputting .33dBm into one of the four output ports and the resulting signal at the input port is -9.23dBm.

View attachment 76740

The theoretical loss of a 4-way divider or combiner is 9dB plus whatever internal loss it has and this divider is working very well despite all the other ports being unterminated. Being very broad band I believe this splitter is a transformer type, but all splitters will work the same regardless of internal topology.

Got any comments Jon? Is there anything else I can do to show you how a splitter/combiner works? Just let me know!
 

jonwienke

More Info Coming Soon!
Joined
Jul 18, 2014
Messages
13,416
Location
VA
OK, I'm not trying to be an ass here, I'm trying to understand what is going on in combiner mode when inventorying the energy budget. I'm assuming a 4-port device with 20dB isolation between ports on the "many" side (A - D), and percentages referenced hereafter are total energy. I'm calling the common "one side" port S.

4-port splitter mode is pretty easy: 100% in, ~20% out at each port, with ~20% of the energy lost due to inefficiency, therefore signal at ports A-D is 6-7dB below the input level at port S. We agree on that.

But the readings you show in combiner mode are not consistent with a device having 20dB of port isolation. If it had that much port isolation on the "many" side, then the energy budget looks like this:

100% in at port A.
~20% inefficiency loss.
20% out at port S.
1% each out at ports B, C, and D (3% total).

Where does the other 57% of the energy go, and why does it only go there in combiner mode? The readings you got suggest that a lot more than 1% of the input energy is getting from port A to ports B, C, and D in the unit you tested in combiner mode, particularly given the difference in your readings when you terminated B, C, and D.
 

prcguy

Member
Joined
Jun 30, 2006
Messages
15,224
Location
So Cal - Richardson, TX - Tewksbury, MA
Depending on the splitter/combiner type, some of the power will be absorbed by balancing resistors across every pair of output ports in the splitter and the rest will be 100% reflected back to the source, leaving only the 25% of power available in one output of a 4-way combiner. The balancing resistors are put here to smooth out VSWR and minimize ripple when ports are not terminated quite right, but they will absorb some reflected power and they are one of the power limiting factors of a splitter/combiner.

As an experiment you can make a simple splitter with two critical lengths of 75 ohm coax (multiples of 1/4 wavelength I think) and a T adapter and leave out the balancing resistor like a phasing harness for a pair of antennas. Now connect a signal source to the T adapter, one cable to a power meter with load and leave the other cable unterminated. In this case the terminated cable will output about 50% of the power and the unterminated port will have 100% reflected power back to the input. Connect it as a power divider and a similar thing happens, power will go up one leg of the coax, split between the common port and the unterminated cable, then some power goes up the unterminated cable where its 100% reflected, a little of that power might be available at the common port and the rest is reflected back to the source. This type of splitter/combiner is very narrow band but schematically similar to a Wilkinson divider.

Or look at it this way, take a 4-way splitter and split a signal four ways. You will have about 25% power at each output port. Then take four equal length cables and connect those four output ports to another 4-way splitter backwards, combining all the power back. In real life you will have the power you started with before splitting minus a small fraction of a dB for general loss. But if your thinking were true you would get 6dB of gain doing this if each input port of a combiner had zero loss through it as you say, and you then connected four in phase signals to the 4-way combiner. That can't work in the real world.

The port to port isolation in a Wilkinson divider is from the 1/4 wave vs 1/2 paths inside the divider. A single port splitting to two other ports using 1/4 wavelength lines works great but using 1/2 wavelengths will cancel the power splitting or combining some at the design frequency. So you have a single port going to two ports using 1/4 wave or 3/4 wavelength lines which allows the signal to flow just fine but between those two output ports its a 1/2 wavelength or multiple of 1/2 wavelength and the signals cancel some between output ports. This is were you get port to port isolation between output ports.

Port to port isolation doesn't really factor into through loss very much and one splitter/combiner with 30dB isolation between ports will not have much loss advantage over one with 15dB isolation except for that isolation, which is important for certain applications.

Sorry if this isn't clear and if you were here I could draw things out with more detail. But then you would get my cold and we don't want that, so you should stay home.

OK, I'm not trying to be an ass here, I'm trying to understand what is going on in combiner mode when inventorying the energy budget. I'm assuming a 4-port device with 20dB isolation between ports on the "many" side (A - D), and percentages referenced hereafter are total energy. I'm calling the common "one side" port S.

4-port splitter mode is pretty easy: 100% in, ~20% out at each port, with ~20% of the energy lost due to inefficiency, therefore signal at ports A-D is 6-7dB below the input level at port S. We agree on that.

But the readings you show in combiner mode are not consistent with a device having 20dB of port isolation. If it had that much port isolation on the "many" side, then the energy budget looks like this:

100% in at port A.
~20% inefficiency loss.
20% out at port S.
1% each out at ports B, C, and D (3% total).

Where does the other 57% of the energy go, and why does it only go there in combiner mode? The readings you got suggest that a lot more than 1% of the input energy is getting from port A to ports B, C, and D in the unit you tested in combiner mode, particularly given the difference in your readings when you terminated B, C, and D.
 

jonwienke

More Info Coming Soon!
Joined
Jul 18, 2014
Messages
13,416
Location
VA
Or look at it this way, take a 4-way splitter and split a signal four ways. You will have about 25% power at each output port. Then take four equal length cables and connect those four output ports to another 4-way splitter backwards, combining all the power back. In real life you will have the power you started with before splitting minus a small fraction of a dB for general loss. But if your thinking were true you would get 6dB of gain doing this if each input port of a combiner had zero loss through it as you say, and you then connected four in phase signals to the 4-way combiner. That can't work in the real world.
I never said you'd get more energy at the S port than what goes in at A-D. My understanding was that with a transformer-type combiner, that if 100% of your energy goes in to port A, that approximately 80% went to port S, and the rest was lost due to inefficiency and leakage to B, C, and D. That the windings of the transformer(s) were wired such that power applied to A canceled out (mostly) at B-D, but was in-phase at S.

But you're saying resistors in the circuit dissipate the "missing" 57% of the energy in combiner mode when you do that.
 
Last edited:

prcguy

Member
Joined
Jun 30, 2006
Messages
15,224
Location
So Cal - Richardson, TX - Tewksbury, MA
The balancing resistors will dissipate some power under high VSWR conditions but I don't know exactly how much. I've blown up many balancing resistors over the years running combiners way out of spec and just a few weeks ago one went up like an M80 on an exposed high power combiner in a 400w UHF amplifier that I accidentally went too far out of band on. If you take the resistors out the splitter will still work but with some specs degraded and it will still have a similar through loss in splitter or combiner mode.

This has gone a bit off topic I suppose and I have nothing else to add for the OP so I'm done here.

I never said you'd get more energy at the S port than what goes in at A-D. My understanding was that with a transformer-type combiner, that if 100% of your energy goes in to port A, that approximately 80% went to port S, and the rest was lost due to inefficiency and leakage to B, C, and D. That the windings of the transformer(s) were wired such that power applied to A canceled out (mostly) at B-D, but was in-phase at S.

But you're saying resistors in the circuit dissipate the "missing" 57% of the energy in combiner mode.
 

TailGator911

Silent Key/KF4ANC
Joined
Feb 12, 2005
Messages
2,687
Location
Fairborn, OH
All the hoopla and nitpicking aside, invest the money in a Stridsberg and be done with it. They are the best multi-couplers made in the industry. You can spend the money on CATV splitters and other after-market couplers, and you will spend more money and time trying to figure out your signal loss than if you invest one time with a Stridsberg. Stridsberg is the best hands down.
 

dlwtrunked

Member
Premium Subscriber
Joined
Dec 19, 2002
Messages
2,109
Any amplifier will increase noise slightly. With a good unit like the Stridsberg, it won't be noticeable unless you are measuring with lab equipment.
Caution here...If the amplifier has a better noise figure than the receiver and enough gain, the result can be a better signal-to-noise ratio (Frii's Equation) using it and better reception. It is signal-to-noise ratio that counts--not just noise.
 

jonwienke

More Info Coming Soon!
Joined
Jul 18, 2014
Messages
13,416
Location
VA
True. The Stridsbergs don't have much gain though--they mostly amplify just enough to compensate for the reduction from splitting the signal. Gain spec is -1 to +2dB throughout the freq range.
 

prcguy

Member
Joined
Jun 30, 2006
Messages
15,224
Location
So Cal - Richardson, TX - Tewksbury, MA
True but Stridsberg doesn't publish their noise figure spec. All they are is a tiny $.50 amplifier chip on a circuit board before the splitter. Their IP1 at 17dBm and IP3 at 32dBm is not that great and will create lots of IMD at my location. In So Cal here we need at least 10dB better specs. That's why I roll my own multicouplers and distribution systems using very high level preamps with reasonable noise figure and IP1 of 26dBm and an IP3 of 46dBm. That seems to work great here feeding a MiniCircuits splitter with better port to port isolation over a Stridsberg and no hint of overload or IMD.

Caution here...If the amplifier has a better noise figure than the receiver and enough gain, the result can be a better signal-to-noise ratio (Frii's Equation) using it and better reception. It is signal-to-noise ratio that counts--not just noise.
 

SDEC_SCAN

Member
Joined
Jun 23, 2015
Messages
6
I can confirm this works. I did notice some signal loss and more digitalization so i hooked up a pre-amp and everything is crystal clear now.
 

Ubbe

Member
Joined
Sep 8, 2006
Messages
8,944
Location
Stockholm, Sweden
Stridsberg overloads easily and the noise figure are not great and the cost for one doesn't reflect its performance. But it is an easy plug&play unit for those who have the money to spare and doesn't need low noise devices and doesn't have interference problems. But I must say that their customer support are excellent.

If I connect a low noise amplifier to the input of my stridsberg and attenuate the signal I get noticable lower noise in the weak signals I receive. So the noise figure in the stridsberg are on par with a scanners noise figure, which isn't great in any scanner.

/Ubbe
 

TailGator911

Silent Key/KF4ANC
Joined
Feb 12, 2005
Messages
2,687
Location
Fairborn, OH
But I must say that their customer support are excellent.
/Ubbe

Ditto on that Stridsberg customer service. I have 5 of their couplers, used them for a few years now, and one went bad. Called them and spoke to a very friendly old ham who said he would send me a shipping label. Two days later I got it, in the package with a brand new coupler. Used the same container and shipped the bad one back to him. Fast, easy, excellent customer service.
 

MTScannerNut

Member
Joined
Aug 25, 2017
Messages
135
Location
NW Montana
Stridsberg overloads easily and the noise figure are not great and the cost for one doesn't reflect its performance. But it is an easy plug&play unit for those who have the money to spare and doesn't need low noise devices and doesn't have interference problems. But I must say that their customer support are excellent.

If I connect a low noise amplifier to the input of my stridsberg and attenuate the signal I get noticable lower noise in the weak signals I receive. So the noise figure in the stridsberg are on par with a scanners noise figure, which isn't great in any scanner.

/Ubbe

Are their any "off the shelf" options for an active multicoupler that is superior to Stridsberg? I'm wanting to split my Diamond discone signal into two scanners. I've read posts about making your own multicouplers, but I don't think I have the skills to do it. Rather buy one ready to go.
 

MTScannerNut

Member
Joined
Aug 25, 2017
Messages
135
Location
NW Montana
A friend recommended this brand to me, but after I had already purchased the Stridsberg. He swears by this one and I thought I might try one if and when any of my other multi-couplers head down south. Good reviews and a bit more affordable than the Stridsbergs.


Thanks for the link. I wonder how the specs on the Cross Country Wireless multicouplers compare to Stridsberg?
 

prcguy

Member
Joined
Jun 30, 2006
Messages
15,224
Location
So Cal - Richardson, TX - Tewksbury, MA
The IP3 specs are better on the Cross Country Wireless VHF/UHF version at 40dBm compared to the Stridsberg at 32dBm. This means the Cross Country will work in dense RF environments better than the Stridsberg. We still don't know the noise figure of either unit or how much gain/loss per port the Cross Country has.

If I was going to buy something like this I would contact both mfrs to get the missing info then go from there.

Thanks for the link. I wonder how the specs on the Cross Country Wireless multicouplers compare to Stridsberg?
 
Status
Not open for further replies.
Top