Sidebar Sidebar
  • You can now help support WorldwideDX when you shop on Amazon at no additional cost to you! Simply follow this Shop on Amazon link first and a portion of any purchase is sent to WorldwideDX to help with site costs.

high swr when more power is added

shockwave, thanks so very much for the lesson.

booty monster and i are both on a continual search for the "real" answers, and this is something i guess i dont understand as well as i thought i did.

if i understand what you are saying, i am now thinking that my DX400 (4x2290)
amplifier is unstable in its highest power setting, causing oscillation (does this have anything to do with drive level? ie: saturation?)

so, if thats the case, what can be done to the average texas star style amp (or any CB amp for that matter) to "stabilize" them?
aside from adding negative feedback that is. (texas stars already have it)

great insights here, and i hope you will expound upon what you have already posted.
this kind of knowledge is sorely lacking in the CB world and it is causing all kinds of head scratching and arguments.
LC
 
What does that SWR meter tell you, what is it doing? It's comparing the impedance it sees at it's input side with the impedance it see at it's output side. (Actually, it compares the voltage or current relation between the two sides of the meter, which will correspond to the impedance.) That SWR meter is constructed to deal with 50 ohms impedance. If the impedance at the input side of the meter is different than 50 ohms then the meter will give an erroneous reading. You work through the circuit of an SWR meter and see if that's true or not. You won't believe it till you do, so do it.
The next thing is that SWR meter can only make gross readings of impedance, it has no idea what impedance consists of (resistance and reactances both (+) and (-)). There are a large number of combinations of resistance and reactances that will seem to be 50 ohms to that SWR meter when the 'R' is NOT 50 ohms. And, because of -where- that SWR meter is placed to get a reading, the feed line and the antenna's combined impedance is whats being read, NOT just the antenna's impedance. That's why the length of feed line can made a big difference IF the antenna's (load's) impedance isn't "50R +/- 0J". If the load's impedance is really "50R +/- 0J" then the length of the feed line isn't pertinent at all if it's characteristic impedance is '50 ohms'. The 'biggy' in the whole mess is that "J" (or "X" for the typical antenna analyzer). The "R" value os impedance is what determines power, that "J" or "X" doesn't contribute anything to power production other than to reduce/change it if present. Drag that 'R' value back to plug into the power equation, see what happens. That power is what's being radiated from the antenna. This is all AC remember, and reactances do not contribute to power production, only the "R" does.
Until YOU do that 'working through' stuff and prove it to yourself, you aren't going to take my word for it. So, prove me wrong. Or right! See for yourself. DON'T take my word for it, or anyone else's, prove it to yourself.
If an explanation of something has to go through some weird, contorted thinking to get to a solution, you are probably looking for the answer to a problem in the wrong place. If you know the load is right, and the SWR changes, it's a pretty fair bet that the load isn't the problem, so look for the problem each 'step' back up the line till you find it. If the load is right and the feed line is right, then what's the next 'step' back up the line? It's either the meter or the jumper between meter and amplifier, or the amplifier. Assuming that the meter is decent (not perfect, just decent), and the jumper is of 50 ohm characteristic impedance, where does that leave you? If that amplifier is a transistor amplifier which uses a common "no tune" output impedance circuit, I'm certainly willing to bet that's where the 'problem' is. It's a 'dead nutz' common problem with 'one size fit's all', 'no tune' circuits. That's also why a 'tube' type amplifier seems to 'cure' the problem. It doesn't have a "no tune" output circuit, you have to tune/adjust it for best output, which also means matching the impedance of the antenna system.
Hell of'a lot of trouble just so you don't have to twiddle a couple of knobs, ain't it?
- 'Doc
 
  • Like
Reactions: Slowmover and troyota
w5lz (doc) you make alot of sense there for sure....... i have had some experience with the input and output impedance differences and problems that can occur with them. im not going to say who is right or wrong. i am just thrilled that i have gotten so many great responses and help in finding the problem maker in my situation. as for now we have narrowed it down to the transistor amp. at this point i dont know for sure the cause of the problem but i have 2 excellent places to look for the trouble from 2 very helpful people. i just want to say that i would not even know where to look if it wasnt for your guys help. and for that i thank you very much
troyota
 
W5LZ said:
What does that SWR meter tell you, what is it doing? It's comparing the impedance it sees at it's input side with the impedance it see at it's output side. (Actually, it compares the voltage or current relation between the two sides of the meter, which will correspond to the impedance.) That SWR meter is constructed to deal with 50 ohms impedance. If the impedance at the input side of the meter is different than 50 ohms then the meter will give an erroneous reading. You work through the circuit of an SWR meter and see if that's true or not. You won't believe it till you do, so do it.
The next thing is that SWR meter can only make gross readings of impedance, it has no idea what impedance consists of (resistance and reactances both (+) and (-)). There are a large number of combinations of resistance and reactances that will seem to be 50 ohms to that SWR meter when the 'R' is NOT 50 ohms. And, because of -where- that SWR meter is placed to get a reading, the feed line and the antenna's combined impedance is whats being read, NOT just the antenna's impedance. That's why the length of feed line can made a big difference IF the antenna's (load's) impedance isn't "50R +/- 0J". If the load's impedance is really "50R +/- 0J" then the length of the feed line isn't pertinent at all if it's characteristic impedance is '50 ohms'. The 'biggy' in the whole mess is that "J" (or "X" for the typical antenna analyzer). The "R" value os impedance is what determines power, that "J" or "X" doesn't contribute anything to power production other than to reduce/change it if present. Drag that 'R' value back to plug into the power equation, see what happens. That power is what's being radiated from the antenna. This is all AC remember, and reactances do not contribute to power production, only the "R" does.
Until YOU do that 'working through' stuff and prove it to yourself, you aren't going to take my word for it. So, prove me wrong. Or right! See for yourself. DON'T take my word for it, or anyone else's, prove it to yourself.
If an explanation of something has to go through some weird, contorted thinking to get to a solution, you are probably looking for the answer to a problem in the wrong place. If you know the load is right, and the SWR changes, it's a pretty fair bet that the load isn't the problem, so look for the problem each 'step' back up the line till you find it. If the load is right and the feed line is right, then what's the next 'step' back up the line? It's either the meter or the jumper between meter and amplifier, or the amplifier. Assuming that the meter is decent (not perfect, just decent), and the jumper is of 50 ohm characteristic impedance, where does that leave you? If that amplifier is a transistor amplifier which uses a common "no tune" output impedance circuit, I'm certainly willing to bet that's where the 'problem' is. It's a 'dead nutz' common problem with 'one size fit's all', 'no tune' circuits. That's also why a 'tube' type amplifier seems to 'cure' the problem. It doesn't have a "no tune" output circuit, you have to tune/adjust it for best output, which also means matching the impedance of the antenna system.
Hell of'a lot of trouble just so you don't have to twiddle a couple of knobs, ain't it?
- 'Doc
Click to expand...

The SWR meter does not compare what it sees on it's input against it's output connector. It compares the ratio between the incident wave and reflected wave. About the only change you'll see if the output impedance of the device in front of the meter was off would be a reduction in the power transfer efficiency. Both on forward and reflected power. Maximum power transfer occurs when transmitter, coax, and antenna are at 50 ohms.

On the other hand I can see your point if the antenna system were generating reflected power. Then the output impedance of the amp may play some bearing on the true accuracy of the SWR reading because it's the load for reflected power. But that accuracy is compromised because of reflected power generated from the antenna system. It's not created by output impedance mismatch in the amp.

If the output transformer in the solid state amp is not converting the transistor collector impedance to the 50 ohm output, you could say there is a mismatch between these components but this in no way will show as a higher SWR on the output of the amp unless this mismatch pushes the amp into an unstable condition.

It's not combinations of resistance and reactance that provide the SWR match. It's when the antenna is tuned in such a way that the capacitive reactance and inductive reactance of the antenna system cancel each other out and appear as a theoretical purely resistive 50 ohm load at the resonant frequency.

What you refer to as a no tune output circuit in the solid state amp requires no tuning because it is a broadband matching transformer. It's matching impedance is controlled by the turns ratio and padder caps used. Solid state amps don't require tuned output circuits, they require individual filters with cutoff frequencies designed for each band.
 
Last edited:
loosecannon said:
shockwave, thanks so very much for the lesson.

booty monster and i are both on a continual search for the "real" answers, and this is something i guess i dont understand as well as i thought i did.

if i understand what you are saying, i am now thinking that my DX400 (4x2290)
amplifier is unstable in its highest power setting, causing oscillation (does this have anything to do with drive level? ie: saturation?)

so, if thats the case, what can be done to the average texas star style amp (or any CB amp for that matter) to "stabilize" them?
aside from adding negative feedback that is. (texas stars already have it)

great insights here, and i hope you will expound upon what you have already posted.
this kind of knowledge is sorely lacking in the CB world and it is causing all kinds of head scratching and arguments.
LC
Click to expand...

You're welcome. My first question would be what are you seeing that makes you think the amp is unstable in high power. Are you seeing significant increase in antenna SWR on high power that might indicate an oscillation? When using high power on SSB do you see signs of a sustained signal between words? If so it's probably not stable. Drive level can effect an oscillation but the problem is within the amp design if see you these symptoms.

What is usually the case is that in lower power modes resistive attenuators are switched in line and this can reduce the likelihood of an oscillation. As you switch to higher power most of the attenuator is removed. Another way to reduce this problem if the amp is able to handle the drive on high power would be to bypass the power level switch and replace it with a tuned input circuit.

One way to improve the ground to the transistors in the event PC board traces were determined to be too inductive would be to place small solder lugs under the two transistor mounting screws and position them so that they can be soldered to both transistor emitter terminals. One on each side of each transistor. Be very careful not to ground out any other terminals in this area. Don't place this amp in a positive ground vehicle after this mod.
 
"The SWR meter does not compare what it sees on it's input against it's output connector. It compares the ratio between the incident wave and reflected wave."

It doesn't do that comparing thingy between it's input and output, really? Then it shouldn't make any difference how you connect that SWR meter to the transmitter, since it's only measuring stuff on one side of the meter. But then, where does that incident wave come from but the input/transmitter side of the meter? And where does that reflected wave go back to, but to the output of the transmitter?

"About the only change you'll see if the output impedance of the device in front of the meter was off would be a reduction in the power transfer efficiency."

I agree, you would see a reduction in efficiency of that transmitter, in relation to it's output power level because of a not good load impedance, it's not seeing a 'good' load impedance that it likes. But that's indicated by that SWR meter showing an increase in SWR. That ratio of incident to reflected power changes, the reflected power goes up, the incident power goes down, meaning the SWR goes up. (You can substitute amplifier for transmitter in that, works the same way, just a different reference point.)
Now, it the transmitter's output impedance is what changes instead of the load's impedance, you will see the exact same thing happen. So how do you tell which changed? By measuring the loads input impedance by some other method. If there was no change in that load's impedance, then guess where that change had to occur!

That SWR meter -DOES- make a comparison of the impedance on the input side and the impedance on the output side of that SWR meter. If the change isn't on one side of it, it has to be on the other side of it. That meter also is only 'good' for 50 ohms impedance systems, it's made that way. All SWR meters are made to be used with a certain characteristic impedance, which in most cases is 50 ohms. Want to use a 50 ohm SWR meter to get measurements from a non-50 ohm impedance? You'd better start doing some impedance transformations at the input and output of that meter. Or, just use a forward/reflected reading watt meter and find the SWR by formula (odd, where have I heard of that happening before?).
- 'Doc
 
All I can say here is that I agree if there is an SWR mismatch shown on the meter then the reflected wave will see the transmitter as the load for that. But, if the load is representing a 50 ohm impedance to the meter it will not show reflected power regardless of the transmitter output impedance.

Stop debating and load a tube amp up at different impedances by using different loading values and redipping the plate. Once the meter is recalibrated you too will see you can't cause a objectionable change in reflected power by varying the output load impedance.

As Bob already pointed out the output impedance match of an amplifier is only going to be matched at a given value of drive, voltage and current. If your theory was correct driving an amplifier on SSB would cause severe shifts in the output SWR when in reality as long as the meter is calibrated for a given power the ratio of reflected changes little throughout the power range and changing output impedances.

The problem with the amplifier described was instability and oscillation. If there were a mismatch between the transistors and output it would not show up as a high SWR after the amp. You would need to measure that reflected power between the transistor and transformer. Not a simple task.
 
AudioShockwav said:
I have stuck this thread at the top for the time being, There is a good discussion going on here with a lot of useful info.
Kudos guys.

73
Jeff
Click to expand...

is there any reason it couldnt be a permanent sticky ?
you dont find this kind of thread/information on any other cb forum .
 
As a newer operator, I would just like to say that I for one greatly appreciate those on here who are generous enough to share their knowledge with us. It's much easier to be able to focus on learning something if you have the right information to start with rather than wading through piles of disinformation found elsewhere.
 
Perhaps a better demonstration showing how amplifier output impedance does not impact reflected power would be using the modern solid state HF rig with an automatically calibrated SWR meter. As Bob mentioned the output can only be matched to 50 ohms at one value of output power. Designers choose this value to be at full rated power to get the best efficiency.

If you were to set the drive level on CW to produce a 100 watt carrier, this would place the output impedance of the amplifier very close to the 50 ohm design goal (assuming your radio is rated at 100 watts). Care to guess how much the output impedance of the amplifier rises to when you drop the carrier to 10 watts?

Output impedance must rise with the reduction in collector current caused by the reduced drive. Using simple ohms law, if a reduction in drive causes current to drop by ten times at the same voltage, the output impedance has gone up 10 times. Yet the automatically calibrated SWR meter looking at the load will remain rock stable throughout the drive range.
 
Then you have 500 ohms out of the transmitter into a 50 ohm antenna. I thought that was really bad? I thought the whole point was matching a 50 ohm output to a 50 ohm load. How does this not wreak havoc with such a difference? What is the "buffer" that makes it possible to run with such a mismatch?
 
from VSWR and displacement
It says "VSWR is not in any way dependent on the equivalent source impedance of the generator."

That is a fact, however in practice it is most likely that if you attempt to change source impedance of your transmitter, you will likely be changing any significant harmonic/spurious content and that will change your VSWR. Unless you are transmitting into a load which has same impedance across all frequencies, which antennas are not.
 
swanny said:
Then you have 500 ohms out of the transmitter into a 50 ohm antenna. I thought that was really bad? I thought the whole point was matching a 50 ohm output to a 50 ohm load. How does this not wreak havoc with such a difference? What is the "buffer" that makes it possible to run with such a mismatch?
Click to expand...

While the mismatch may not seem ideal, you have to realized there are limitations to matching networks. In order to completely correct this type of mismatch it would require something like an impedance regulator that could constantly recorrect the changing impedance. I don't think I've seen this device in any standard radio application.

The reality is these mismatches do not cause a problem since the match is ideal at high power where it matters most. The match slowly gets worse as the as the power is reduced but that's not a problem because the simultaneous reduction in current keeps the transistors dissipation well within specs.
 

dxChat
Help Users
  • No one is chatting at the moment.


      Top Bottom
      Back