• 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.
  • Click here to find out how to win free radios from Retevis!

power output question about the sd1446

I have components to do the negative feedback mod. I havent done anything inside the amp yet though. I will work on getting my equipment properly configured first.

Thanks,

Big D


ok, this is why i always ask these kinds of questions. while i don't see anything in your system that immediately strikes me as a culprit of your amp not working, and that is good news; i do see potential for a safety issue with your set up.

i'll start with the amp first.

looking at the picture you posted, you will see that each 1446 transistor has four gold foil tabs coming off of it.

you will also notice the black ferrite beads with little PC boards on their ends and wires wrapped through them, and how they sort of frame the 1446's.
these are called broadband transformers and the little one's job is to convert the 50 ohms out of your radio into the approx. 2 to 3 ohms that the transistors want to see.

the big broadband transformer's job? you guessed it,it converts the 2 to 3 ohms from the transistors and converts it back to 50 ohms.

so, looking again at the four tabs coming out of each 1446, the tab that connects to the small broadband transformer is the BASE of the transistor.
the tab that connects to the big broadband transformer is the COLLECTOR.
the two tabs that go off to the sides are both EMITTERS.

to install a negative feedback circuit into the amp; first connect one end of the capacitor to one end of the resistor and solder them together. like this: --------/\/\/\/\------[(------
now do the same thing again so you have two of them.

now solder the capacitor side to the base of a transistor, and the resistor side to the collector. do it for each 1446.
you want the cap and the resistor to float and not be touching anything else, but you also want to keep the leads as short as possible, so take some time and fit things before you solder them in place.

i was able to find some info on an amp that should be pretty much identical to yours:
http://www.cbtricks.com/Amp/palomar_other/blue_lightning/index.htm

this page should have all the info you need once you learn what is what.
first look at the "PCB Layout" and you will see that it pretty much matches the pic you posted. this would mean that C22 is the input cap we were talking about before.
the part number may be different in your amp, but it will be located in the same place.
tell me what value this cap is in your amp.

Back to the grounding system.
The disclaimer here is that i don't know any of the codes where you live, and i am not a certified electrician, so anything i say here should be checked with other sources more in the know before you take action.

the problem with grounding your radio equipment really well, but not tying it in to your main electrical system ground is that if the mains ground isn't very good, once you connect your antenna coax to your radio, you might now be providing a better path to ground than your mains system, and its running right through your radio equipment.

this could mean that when you turn your computer on in another room, that it's path to ground goes right down your antenna coax, through the mast, and to your ground rod.

this can be dangerous for lightning, and can also cause weird problems with RF equipment.

the cure is to run a large solid copper wire from your antenna ground rod a few inches under the ground, and connect it to the main ground for the house electrical system.
This is in the NEC code book.

ideally, you would also install a ground rod right outside your shack within 7 feet or less of your operating desk.
each piece of equipment would be run to a copper panel in the wall or the back of the desk, and a large copper wire would run from that panel to your "shack" ground rod.
the wire that connects your station to this ground rod should be less than 8 feet long.
This "shack" ground rod would also be tied to your other two grounds so that everything is at the same potential.

im saying all this to give you something to look at.
i am by no means trying to take control of your station and tell you how to run it.
your amp problems are probably not stemming from anything to do with the antenna system.
LC
 
Big D,

i was just looking at the picture of the amp you posted, and wanted to suggest that you take a little time to remount and resolder some of the more sloppily placed components in your amp before you go too much further with your experimenting.
for example, C8, that little ceramic disc cap right off of the input broadband transformer, should be mounted right up against the PC board with as little lead length as you can get.
you can see how they mounted it sloppily leaving the leads really long.
this can mess with the input tuning of your amp and should be corrected.
do this anywhere you find this type of sloppy soldering and mounting.
this stuff matters when it comes to RF!
LC
 
C22 is labeled a 103 it's value is a .01uf.

I will try to look into grounding antenna to house ground (Easy Job). House uses cold water line for ground. May take me a little time but also get radio equipment on a second grounding rod and get all grounds tied together as well.

Thanks for the help (-:

Big D
 
no problem.

If after you install the negative feedback circuit, you still see the high SWR going in to the amp, the next step would be to try a different value cap in that location.

first try a .022uF ceramic disc cap. use one with a voltage rating of 100v or more.
If that only makes things worse, try a .0047uF ceramic disc cap, also with a 100v+ rating.

If neither of those two values changes anything significantly, then the next step is to try the input tuning mod. put the stock value cap back in C22 before continuing.

looking at your amp, you can see that there is a wire wrapped around the input broadband transformer (the smaller one), and one end of this wire connects with one side of C8.
to add the input tuning network you would lift this wire from the PC board and solder in your 5 turn, 16ga. solid copper wire coil that you wound around a 1/2 inch form. (drill bits are good for this) the coil turns should be right next to eachother, not spaced apart.

the way to count turns on a coil is to first make a right angle bend in one end of your wire, and count this point as the beginning of turn number 1.
place the wire on your drill bit or other form so that the short bent end of the wire is parallel with the bit, and the long part of the wire is sticking straight up in the air.
now tape or clamp the short wire in place on the bit to keep everything steady while winding.
now take the vertical part of the wire and start tightly coiling it around the form.
when your first turn meets up with the bend in the wire, that is one turn.
wrap it around again and when it meets up with the bend again, that is two turns.
continue on until your fifth turn meets up with the bend, and then make a right angle bend in the wire at this point.
your two bent ends of your coil should be along the same plane.

now snip off the ends of your coil until the bends at the ends are about 1/2" long or so.
(not critical, just easier to mount). strip the wire all the way back to the bends.

solder one end of your new coil in the hole vacated by the broadband transformer wire,
and solder the other end to the wire of the broadband transformer.
be sure to leave a little bit of bare wire sticking up out of the hole to give you some room to solder your trimmer capacitor to the coil. leave some bare wire at the other end too.

The other end of the wire that is wrapped around the broadband transformer goes to ground.
unsolder this wire and solder a short length (about 1/2") of bare copper wire to the now empty hole. solder the broadband transformer wire to the end of the bare copper wire sticking up. what you are doing here is creating some room to solder your trimmer capacitor to ground, and to the PC board end of your coil.
an easy way to mount the trimmer cap is to cut a couple of pieces of solid copper wire about an inch long, and bend small hooks in each end.
hook one end around each side of the trimmer capacitor and solder in place.
yes, all the little metal "wafers" get soldered together.
now you have a trimmer capacitor with little wire nubs sticking out each end with little hooks bent in them.
hook one end around that little piece of wire sticking up out of the PC board that connects to ground and the broadband transformer wire, and solder in place.
hook the other end around the PC board end of your coil and solder in place.

it should be pretty obvious now that we are trying to make the trimmer cap accessible enough so you can tune it without touching anything else in the amp.

There is one more point to make about the trimmer capacitors. if you get your ohm meter out, and set it to measure continuity, you will notice that one end of the trimmer is connected to the little screw head, and the other side is not.
the side that IS connected to the screw head should always go to ground.
this way you won't mess with the tuning with your screwdriver.

now hook the amp up with the SWR meter going in to the amp, and check the SWR.
start making half turns on the trimmer trying to lower the SWR.
you will need to recalibrate the SWR meter each time you make a change to the trimmer capacitor.

you should notice a pronounced "dip" in the SWR at a certain position of the trimmer.
if this "dip" occurs, but does not dip low enough, try spreading all the turns in the coil just a little bit. try to space each turn evenly. im talking about 1/16" spacing to start with.
try your tuning again, and remember to recalibrate every time you make any change to the system.

lets stop here for now, and give you a chance to do some experimenting.
be sure to post back with the details of what you encountered at each step along the way.


as for the station grounding, there are many good articles on line about this, but i would stick to the ones about ham radio stations. some of the CB grounding articles have a bit of snake oil in them sometimes.
yes, it will get confusing.
the bottom line is to make sure all the grounds are tied together, all your connections are bright and shiny, and then coated to protect them from weather.

when it comes to the station ground. specifically the wire that runs from behind your desk to the ground rod outside the shack, just try to avoid any lengths between 7 and 10 feet.
like choose 6 feet, and if thats not enough, step up to 11 feet.
if 11 feet isnt enough, stay away from 14-20 feet, and go with 21 feet.
this wire should really be kept as short as possible, and if you are looking at a 20+ foot long ground wire, you might want to consider another location or option.

best of luck, and be sure to post back with your findings.
LC
 
Last edited:
I'm not gonna say it is the issue with this amp. But I have come across a lot of amps that weren't tuned right. Not just changing the value on the input and output. But the actual value across the transformers as well might have to be changed. LC can go into more detail if need be. But I'm pretty sure he knows where I'm going with this.
 
I sure do Crusher, and i really hope that's not the case.
I have been down that road before with a homebrewed amp, and i have to admit that i never did find that balance between low SWR and full power out.

there are just too many things that effect other things, and my knowledge of the correct equations and relationships to use just isn't up to par to be able to give conclusive advice.

if that ends up being the case, and Big D wants to continue on, i might have to hand the reigns over to you real amp gurus.

im just hoping i haven't given any bad info so far! LOL
LC
 
Your doing good so far. Jumpers, coax, ground and match are the first things to check. Next is a few different radios. If they are butchered or an improperly tuned export, good luck with that. Next is the amp itself. Lol, that's when it's time to stock up on 465's and something like 4610 or 4615's.
 
From the word go we know the efficiency of this amp can not be that high!! It is a transformer coupled design so get rid of those huge number's! Your lucky to get 250 watts rms out of a TS667 on AM let alone 2x SD1446's. If the amp is working right than the meter is not at all accurate. No big deal but you can not trust it if that amp is working as it should be. So at this point you need to decide what direction to take. We either assume the meter is bad and move on or we assume the amp has an issue and we need to figure out what it is. Unlike the Toshiba 2SC2879's and a lot of MRF transistors the SD1446's gain was never really under rated and you seldom got much more out of them by volting them or continuing to add more drive beyond spec.! Have you considered using a VOM and checking out put voltage and comparing on a chart to determine true output power to check against your meter? No way on this Earth 2 watts will give you 250 out even on high unless that unit is oscillating or the meter is way out of calibration and the diodes inside have gone south of their rating. Voltage is taken into account when you run the gain number's versus input and output. As voltage climbs so does gain until you reach the saturation point that is why some transistors that have a lot more headroom for lack of a better word are able to to see increased power with increased voltage. Basically the engineers built in a larger margin for error to increase durability so that at the designs worst given quality control of the day it would still perform up to the white papers minimums all the time. Not every transistor is that over built or under rated. The math does not add up no matter how you look at it!!
 
I have a "palomar 35a" which has 4 X SD1446 and I get 300-350 out of it...thats what it should do. If you have a 2x 1446, I would say 150 should be normal. ..especially from only a 2 pill amp.
 
  • Like
Reactions: Captain Kilowatt
WOW! I enjoyed reading through this old post. Perhaps someone should explain why negatrive feedback is employed and why all amplifiers do not need to employ it if other design parameters of the amplifier are considered and it use is limitied to a specific band.
 
  • Like
Reactions: Shadetree Mechanic
Then perhaps a study in the definition is needed...

upload_2021-1-23_17-32-4.png

You can go here...
https://en.wikipedia.org/wiki/Negative_feedback

And read all about the basic concepts...

But what is being talked about here, is a form of error correction - in the performance of a circuit and to keep it stable over a wide range of power levels and input fluctuations as well as the effects of changes exerted on the output.

So, if you're designed an amplifier but need to keep it stable, you build feedback into it - by tuning or adjustment as a from of observation and control - a form of ERROR CORRECTION..

So in a way, you already answered your own question...
why all amplifiers do not need to employ it if other design parameters of the amplifier are considered and it use is limitied to a specific band.

Specific band is a main requirement

In nearly all aspects - if you design the amplifier as a unit, to perform one simple duty, the "feedback" effort is placed in the unit as a form of it's function - input level range? - SET, output level gain? - SET -then the feedback itself is in the performance of the unit.

Then why the extra parts as a function of the loop?

Look at the Datasheet, since the 1446 isn't the best at showing the complexity of the matching you need for every given input power levels to the results of the output - perhaps a better idea to look at this from the standard of a performance hit regarding distortion.
  • The ability to be faithful to reproduce what is present at the input - into a larger signal is not guaranteed so you will have problems, called distortion. They are products that are the end result of the amplification.
  • Some devices are better and provide less distortion when properly driven and need very little correction.
  • The SD1446 - if properly driven can accomplish this without the need for feedback for the feedback paths to achieve this were already incorporated into the circuit to show the devices ability to produce power under a given set of criteria.
When you send a signal into this unit, if it's properly set level and not beyond a set range of values - like Power input - you would not need to have to worry about the amplifier stages ability to develop the power into a larger value or gain result, because if all things are properly set - no other action needs to be taken. You then can define this a a closed loop of gain - not an open loop of gain which many amplifiers can be though of and built as.
.
Same can be said for the output - if you know where it is going, and what values that it will have - you again, would not need to worry about some form of error correction if you did your job and installed this unit correctly. Again, Open loop gain doesn't exist - the stages gain was already considered and values are set.

Using negative feedback, is a type of PROCESS needed for amplifiers of stages that are designed with open loop gains in mind.

Why? Because - by asking that question - you're forgetting about the simple changes you have to accomplish using a complex set of parts to achieve the proper input levels for the device (Your SD1446) in the unit.

Look at post #83 - kinda ties a ribbon on this for the sake of knowing that what you have to do to an RF signal is de-convert it from a 50 ohm unbalanced state - into a working low-impedance input value to enter into a device that gives you another type of complex output that in itself requires a conversion to return this signal - now amplified - into the 50 ohm, in a method so the system it's installed in, can use it.

So - if you can envision that input and output Transformer, or; "tubes" Trombones, Chokes, Coils, Pipes or what ever the poster may have called this separate UNIT within this Unit - you can now begin to see a set of problems with both complexity and limitations of it's ability - you also can begin to see a set of units within a unit itself that are required so the unit - as a whole - can function as an amplifier.

It's best described in this paragraph from the WIKI-;
upload_2021-1-23_17-55-8.png

It's important to remember that the UNIT being installed, even if done correctly - the keyword being Correctly means that some form of feedback has already occurred - a test.

It was tried (in many ways) in a condition. Whether it passed or not - isn't the condition - the passing of the test means that a type of feedback already was done to the UNIT to make it work as a device to drop in - it means a result was obtained in a consistent fashion .SEE: Tuning​

When you say Negative Feedback - the purpose of that, was for the Unit within the Unit - a type of control against an error in it's output that is possible because of the units around it can impose changes on it.

Changes that are not part of the original design - to amplify - so if a input signal is too strong - it can cause an error in the ability to produce an output faithfully to it's input.

Same can be applied to the output of the Unit within the Unit - if the antenna or the coax changes, then the output is in jeopardy of now being wrong or possibly the output of the Unit can be damaged by the losses incurred from those changes in the Unit that is responsible for for the complex conjugate to bring the power developed, as a different impedance, back to proper impedance as it's output.

So to me, IMHO and all other disclaimers apply, the effort of the Negative Feedback, is a realistic application of a feedback process to keep the main unit of the amplifier working correctly as much as possible when all other variables to the input and output cannot be controlled

Which is-
  • the environment it operates in,
versus the;
  • - environment the units of the main UNIT work,
    • it's basic function to provide an operation, the amplification
  • These units - which are also fixed and set but allow for this, or these condition(s) to occur
    • can't act upon the error - but provide a WINDOW or RANGE of level so the Unit can process the signal to a degree of performance
- so the best case scenario is satisfied by the efforts of the smaller conversion units being able to accommodate the external environment being applied, or operated in, as a means to isolate the amplifier within main Unit from the external so it can simply amplify what is there on it's input. Your reference to Design Parameters as elements within the Unit itself.
So if you can think of the amp as a set of smaller blocks that function to help the unit obtain this "figure of merit" the negative feedback issue becomes more of a "does this require more control" event. Which in many cases of mobile setups - it's a necessity - to keep the Unit stable - you simply cannot control the End user - or the conditions the Unit will work in. Because of the End users choices for the environment it will be used in.

You may wonder Why so much "talk" - well you did ask the question and not many understand the need for a simple Resistor and Cap combination - for the Amplifier to stay stable and reduce it's ability to make errors - it's a requirement so the amplifier won't be damaged or self destruct because of the environment the person is using it in.

To claim it (Feedback) can be removed is a misnomer, the stage has already been adapted to perform, but the open loop gain needs to be addressed.

If you get rid of the requirements to adapt to the environment by lessening the gain demands - the need for feedback simply becomes an effort to make the amplifier a stage of set gain working in the environment of the RF levels of the radio and no need to convert impedance - its' done as a factor of output to convert to the means of connecting it to the outside world.

So don't get the term Negative Feedback, and Open Loop Gain, and Set Gain amalgamated together - are all factors of merit we attempt to obtain but are concepts to the end results.

So the use of Negative Feedback in these cases - is used as both Error Correction, and Gain control.
 
Last edited:
  • Like
Reactions: Shadetree Mechanic
Unfortunaly that's only possible if your amplifier is unstable and oscillating. This transistor is only rated for 10db gain. That means whatever drive you give it, the transistor will try and make 10 times that as output. With the 2 watt carrier the amp should only be keying about 20 watts.

For it to be over 200 watts output with 2 watts drive means the stage would have to have 10 times the gain it actually does or the more likely possibility is it's self oscillating. Check you SWR when it says this power and there is a strong change they will be high. Another key sign something is wrong.
That’s dead key not peak swing. Yes that’s correct. My amplifier dead keys 60 with a 2 watt dead key & peaks 250 with 30 pep of drive. That’s with a pair of SD1446.
 
i have a 2x sd1446 palomar amp or similar, i know the sd1446 are rated at 75 watts. my meter shows it keying up with a 2 watt carrier going in at 80, 150 , 250 watts for low med high power, is this possible? i fairly certain the meter is accurate at least at the lower readings because i know my radio is set to 2 watts on low power and its is right on the reading. ( not to say thats an accurate way to judge my meter buts its all i have)
They are rated @90 watts each
 
i have a 2x sd1446 palomar amp or similar, i know the sd1446 are rated at 75 watts. my meter shows it keying up with a 2 watt carrier going in at 80, 150 , 250 watts for low med high power, is this possible? i fairly certain the meter is accurate at least at the lower readings because i know my radio is set to 2 watts on low power and its is right on the reading. ( not to say thats an accurate way to judge my meter buts its all i have)
90 watt
 

dxChat
Help Users
  • No one is chatting at the moment.
  • dxBot:
    Greg T has left the room.
  • @ BJ radionut:
    EVAN/Crawdad :love: ...runna pile-up on 6m SSB(y) W4AXW in the air
    +1
  • @ Crawdad:
    One of the few times my tiny station gets heard on 6m!:D
  • @ Galanary:
    anyone out here familiar with the Icom IC-7300 mods