Strange...

[T23]

This is strange, I have an HF setup with a 2 pill amp inline and unless the amp gets a certain amount of drive from the radio the SWR readings are 3:1 or greater, but with a cetain amount of drive from the radio SWR readings are normal as if the radio only was conected 1.1:1.or 1.2:1.

Now from what I read in the data sheets about the pills in that amp eatch pill needs an average of 4 watts of drive to be comfortable, or atleast that's what I have gatherd from the data sheets.

My setup:

Galaxy DX 44v

Texas star skeleton 100

I am using a homebrew 1/2 wave loaded vertical antenna with a matching network.

Its tuned for 26.915.


T23

 

[eagle1911]

The SWR at the antenna is 3:1 or higher unless the amp is kicking out a good amount of juice? Or are you reading the SWR at the amp input?

 

[AudioShockwav]

The SK 100 is a bare bones Amp with point to point wiring and has not been made in a long time...
Look at the thread I posted at the bottom.
Read through the thread, there is great info that you will find from Shockwav and Bob 85...when you are done you will have a better understanding of what is going on inside your little amp.

http://www.worldwidedx.com/cb-antennas/56747-high-swr-when-more-power-added.html

73
Jeff

 

[T23]

I see what you mean, so what's happening is that the amp is oscilating and the change in the drive input lowers the SWR because it drowns this oscilation out? I believe the board you were talking about is there, the amp looks all original when I looked at the guts.

T23

 

[eagle1911]

I must be confused AudioShockwav, as it sounds to me like he's describing the exact opposite effect, with the swr being highest at the lowest power output, and lower with more power. If that's actually the case, then I doubt it's harmonic content..

More power will never "drown out" harmonic oscillations. The more power a transistor amp produces (ie the higher the drive), the more powerful the harmonics also become. In push-pull amps the 2nd harmonic is naturally suppressed due to the nature of the design, but the 3rd harmonic can be quite powerful.

 

[T23]

Eagle, I measerd the SWR at the input, the SWR is the same as with only the radio (flat). If a certain amount of drive to the amp is present. Too little or too to much drive and the SWR goes 3.0:1 or higher.

T23

 

[eagle1911]

hmm.. that seems odd to me as well. Typically when harmonics are measurable, they are lowest when the amp is being driven lightly, and highest when the amp is being driven hardest. They certainly can cause higher SWR readings with certain meters, but it's wierd that you are reading these higher SWR numbers even when the amp is being driven lightly.. it's almost like there's a "sweet spot" of some sort.. not sure what could cause that though.

 

[AudioShockwav]

I do not think it has negative feedback in that little amp, and as in the example it is unstable, that, and the "ugly Bug " wiring is adding to the problem.

"

You may open this amp and find the negative feedback is already there since this is not the only cause of this type of oscillation. You'll have to look and see. Poor RF layout and PC boards with inadequate ground surfaces can also be the source of this problem. It is essential that the emitter terminal of the transistors be at absolute RF and DC ground potential.

"

It's important to note that the symptoms described concerning high SWR on the solid state amp are not related to 2nd and 3rd order harmonics where you would need a spectrum analyzer to confirm their level of amplitude or need to filter them out to correct this problem. When an unstable amplifier breaks into oscillation as is the case here, the undesired frequency component is usually not very far from the fundamental. Often within a few MHz. You can buy a cheap scope on eBay for $75 that will spot this problem.

For example, when an amplifier is oscillating look at the carrier wave on the scope. You will see at least two signals mixed together. Lets say the main carrier is at 4 divisions peak to peak on the screen. Underneath that carrier you will see another at perhaps 2 divisions on the screen. The upper and lower envelope of the second undesired signal will often be lower in amplitude and full of distortion within it's envelope. Under conditions of low drive or SSB use, the oscillation may be the dominant signal.

So one may be asking, "Why does adding a TVI filter sometimes reduce the SWR in a case like this? The answer is simple. When you are dealing with an unstable amplifier, any slight change in the inductive or capacitive reactance of the load can cause changes in it's operation. Sometimes even on the input side. You can often find a situation where just adding or subtracting some coax will find a condition where the amp runs stable for now. Adding the TVI filter with all of it's inductors and capacitors can certainly shift these conditions.

" When you are dealing with an unstable amplifier, any slight change in the inductive or capacitive reactance of the load can cause changes in it's operation. Sometimes even on the input side. "

Loose cannon asks "is this because of an impedance mismatch having to do with the drive level of my radio?"

Bob 85"
amplifiers can only ever be 50ohms at one drive level/supply voltage, you having calculated output tank circuit requirements would realise that,

you don't get high vswr when you change drive level or supply voltage on a stable amplifier."
And
"the large increase in vswr is likely to be as shockwave stated, an unstable amplifier"

I think T-23 is getting it...

Shockwave

When the SWR goes from 1.1:1 bafefoot to 2:1 when you turn a solid state amp on, most of the time that's oscillation. In this case turning the drive up often lowers the SWR because the fundamental frequency drowns out the oscillation.

These guys are wayyyy sharper than me about solid state amps,
Maybe they can get in here and go deeper than this.

73
Jeff

 

[eagle1911]

That all makes perfect sense.. I didn't realize there was no feedback. The EB63 I'm working on right now doesn't have feedback by design, but I'm adding a .01uF ceramic cap on each side each in series with a 50 ohm 5W resistor and a VK200 inductor, the value of which I'll be experimenting with.

I took a look at the inside of the skeleton. You could easily find a way to arrange a .01uF ceramic disk capacitor and a 2-5W 50-60ohm resistor connected as I've drawn in the pic below. Due to the restricted space, you'd probably have to go up or to the side, but there's enough room that you can work the components into that space. Just connect the resistor and cap in series and solder the ends to the metal part on the top of each transformer. There would likely be some improvement in stability just by doing that.. You wouldn't need the inductor since the skeleton is only for around 27mHz.

 

[AudioShockwav]

Great picture. I have seen one of these little amps long ago, but could not remember if that small board over the transistors included feedback, I do know that most of the really small amps T/S were built with that point to point wiring set-up....your picture tells the story.

T 23, try adding the feed back and see if it solves the problem, let us know how it works.


Thanks to guys like Shock and Bob that get on here and share all this good info.

73
jeff

 

[eagle1911]

I can't take credit for the pic, I just googled it up and added the feedback components in paint. I needed an easy way to illustrate where to add feedback. It's such a simple thing to add really..

Those guys are the reason I got turned on to this site, which may be a good or bad thing depending on who you are lol.. there's a lot of good stuff here from them and others like them.

 

[Shockwave]

Another common problem with these amps that contributes to this oscillation is that the emitters of the power output transistors are not given proper low inductance paths to ground. Look at that ground connection opposite of the terminal strip with the long wire going all the way to the back before it ties to a ground lug. Carefully drill a screw hole into the heat sink right near that emitter ground tab and solder a lug right to the tab. Clean the black anodizing off the heat sink under the lug before you screw it down. Don't ram the drill into the transistor when you punch through the heat sink. The drill press is your friend here.

 

[eagle1911]

From that picture it looks like the only ground connection is the one connected to the negative power wire at the bottom so-239..

 

[Shockwave]

I think you'll find one end of the emitters is tied to a ground that is closer. Right under that terminal strip. The fact is one good ground on one of four emitter tabs is not enough. That is a problem with the "dead bug" style amps. Ideally each of the four emitter tabs would be given a direct ground right to a PC board. Adding one more ground lug on that transistor with nothing good on it should be sufficient. Be careful not to break or remove those RF beads on the ground wire going to the input transformer.

 

[TonyV225]

I have to laugh when I see some of these older amplifiers!! Some were really cool and nicely done like the Prides older palomars and a few others and then I see some of these things on the net and think WTF!!?? That thing works or worked!? You would think once power was applied there would be ground loops issues ETC Ive been through that crap it was frustrating. Then you try another amplifier and its as good as it gets.