Re: 10 W linear amplifier: output impedance

Christoph Schwärzler

Allison,

thanks for your detailed reply. I know that dealing with output impedance is rather unusual and close to black magic.

The reason I gave it a thought is as follows: Feeding a source with an impedance of 22 Ohm into a 50 Ohm system leads to power loss of something like 15% or -0.7 dB. This is not a lot and most hams probably wouldn't care. Matters are even more complicated as output impedance typically is frequency dependant, as you mention, too.

However, in the QSX (and most other devices) the output stage is followed by a LPF. Most people are used to the symmetrical form (e.g. C1 =C3, L1 = L2) of a Chebycheff design. This need not be. The symmetry is a result of assuming 50 Ohms at both ends when designing the filter. Many (online) calculators assume this or force the user at least to use equal impedances at both ends. It can be easily done differently, though. It is possible to design a LPF with an input impedance of e.g. 22 Ohms and an output impedance of 50 Ohms. Parts count is the same, cost is the same, it is just that C1 equals no longer C3 and L1 is different from L2.

As switching between bands also switch LPFs, the frequency dependence is easy to cope with. So why not design this into the LPFs and gain 0.7 dB for free?

Just a thought.

Vy 73,
Chris

Re: QCX-SSB Help what do I do now?

Pascal Nguyen <vk2ihl@...>

Thanks anyway for not helping how to program the 328p.
So I keep playing with v1.01d then.

I fed the Si5351 clocks to my CRO, mismatched impedance of course, still got a square on my CRO, that justifies the inputs are  π/4 or π/2

Pascal Nguyen

On 5 Mar 2020, at 2:36 pm, ajparent1/KB1GMX <kb1gmx@...> wrote:

﻿...

QCX modified for SSB if it worked  would be cheap way in.  Seems
I'm still trying to make it work after a year though I was forced to take
a break in the middle.  Early code was terrible, and current wants a
lot of experimental setup and tweaking.  More MPU would be a step
in the direction of easier and better

...

Re: 50 watt Amp IRF 510 Failure

Hans Summers

Hi George

Seeing continuity is correct. But, the "any" in the sentence is superfluous, I will have to get that corrected.

"Now cut, prepare, and solder the wire in the bottom left hole, nearest to Q1. Check for continuity with the DVM, from the middle pin of transistor Q1, to the right hand side of capacitor C7. You should see any continuity here. "

"Now cut, prepare, and solder the wire in the bottom left hole, nearest to Q1. Check for continuity with the DVM, from the middle pin of transistor Q1, to the right hand side of capacitor C7. You should see continuity here. "

73 Hans G0UPL
http://qrp-labs.com

On Sun, Mar 8, 2020 at 7:45 AM George Korper <georgekorper@...> wrote:
I have continuity in that test step on page 17 between Q1 and right side of C7. If it should be no continuity, what's up? I removed the transformer and it is not shorted and I black marked the appropriate wires. The manual needs editing for clarity in that step in my opinion.
I won't proceed, as I  have made a working amp which works but is not capable of practical use on 20 meters.

On this second attempt I exactly followed the manual and am stuck here. So I will wait to hear from Hans.
No sense struggling. I do feel bad abou

On Sat, Mar 7, 2020, 9:41 PM Paul AI4EE <nadie1943@...> wrote:

Open circuit.

On 3/7/2020 6:53 PM, George Korper wrote:
The sentence: "You should see any continuity here."
What does that mean?

Re: QCX 40 NO POWER OUTPUT

Jim Allyn - N7JA

On 3/7/20 3:16 AM, Greg Walters via Groups.Io wrote:
It turns out an individual as offered to check it out for me. If he has no luck, and you are still interested, I'll let you know.

Still interested.  Please let me know if it becomes available.  Thanks.

Re: 50 watt Amp IRF 510 Failure

Ronald Taylor

George, I was looking at the final schematic when I made the last comment. If at the point you are at you still have wires remaining on T1 to be connected there may not be continuity at that step but it will come when all the wires are soldered. You best wait for Hams to clarify. 73... Ron

On Sat, Mar 7, 2020 at 21:58 Ronald Taylor via Groups.Io <wa7gil=gmail.com@groups.io> wrote:
George, at that step you should see continuity between the middle pin of Q1 and the right hand side of C7. Both Q1 and Q2 middle pins (drains) should have continuity to the right hand side of C7 If T2 is wound and installed correctly.

Ron

On Sat, Mar 7, 2020 at 21:45 George Korper <georgekorper@...> wrote:
I have continuity in that test step on page 17 between Q1 and right side of C7. If it should be no continuity, what's up? I removed the transformer and it is not shorted and I black marked the appropriate wires. The manual needs editing for clarity in that step in my opinion.
I won't proceed, as I  have made a working amp which works but is not capable of practical use on 20 meters.

On this second attempt I exactly followed the manual and am stuck here. So I will wait to hear from Hans.
No sense struggling. I do feel bad abou

On Sat, Mar 7, 2020, 9:41 PM Paul AI4EE <nadie1943@...> wrote:

Open circuit.

On 3/7/2020 6:53 PM, George Korper wrote:
The sentence: "You should see any continuity here."
What does that mean?

Re: 50 watt Amp IRF 510 Failure

Ronald Taylor

George, at that step you should see continuity between the middle pin of Q1 and the right hand side of C7. Both Q1 and Q2 middle pins (drains) should have continuity to the right hand side of C7 If T2 is wound and installed correctly.

Ron

On Sat, Mar 7, 2020 at 21:45 George Korper <georgekorper@...> wrote:
I have continuity in that test step on page 17 between Q1 and right side of C7. If it should be no continuity, what's up? I removed the transformer and it is not shorted and I black marked the appropriate wires. The manual needs editing for clarity in that step in my opinion.
I won't proceed, as I  have made a working amp which works but is not capable of practical use on 20 meters.

On this second attempt I exactly followed the manual and am stuck here. So I will wait to hear from Hans.
No sense struggling. I do feel bad abou

On Sat, Mar 7, 2020, 9:41 PM Paul AI4EE <nadie1943@...> wrote:

Open circuit.

On 3/7/2020 6:53 PM, George Korper wrote:
The sentence: "You should see any continuity here."
What does that mean?

Re: 50 watt Amp IRF 510 Failure

George Korper

On Sat, Mar 7, 2020, 10:45 PM George Korper via Groups.Io <georgekorper=gmail.com@groups.io> wrote:
I have continuity in that test step on page 17 between Q1 and right side of C7. If it should be no continuity, what's up? I removed the transformer and it is not shorted and I black marked the appropriate wires. The manual needs editing for clarity in that step in my opinion.
I won't proceed, as I  have made a working amp which works but is not capable of practical use on 20 meters.

On this second attempt I exactly followed the manual and am stuck here. So I will wait to hear from Hans.
No sense struggling. I do feel bad abou

On Sat, Mar 7, 2020, 9:41 PM Paul AI4EE <nadie1943@...> wrote:

Open circuit.

On 3/7/2020 6:53 PM, George Korper wrote:
The sentence: "You should see any continuity here."
What does that mean?

Re: 50 watt Amp IRF 510 Failure

George Korper

I have continuity in that test step on page 17 between Q1 and right side of C7. If it should be no continuity, what's up? I removed the transformer and it is not shorted and I black marked the appropriate wires. The manual needs editing for clarity in that step in my opinion.
I won't proceed, as I  have made a working amp which works but is not capable of practical use on 20 meters.

On this second attempt I exactly followed the manual and am stuck here. So I will wait to hear from Hans.
No sense struggling. I do feel bad abou

On Sat, Mar 7, 2020, 9:41 PM Paul AI4EE <nadie1943@...> wrote:

Open circuit.

On 3/7/2020 6:53 PM, George Korper wrote:
The sentence: "You should see any continuity here."
What does that mean?

Re: 50 watt Amp IRF 510 Failure

Paul AI4EE

Open circuit.

On 3/7/2020 6:53 PM, George Korper wrote:
The sentence: "You should see any continuity here."
What does that mean?

Re: 50 watt Amp IRF 510 Failure

ajparent1/KB1GMX wrote on 3/7/2020 9:06 PM:

I can't see.   I literally have no idea.   I am stabbing in the dark.
Hello--

Hey, I'll take a stab too<g>.

Suppose that the RF output transformer core is mismanufactured or mislabeled?
Standard practice is to color-code the core to designate its properties.

Somewhere in China, a  worker at a magnetic-core factory
ran out of  a certain paint color, shrugged his or her shoulders, and
reached for the next-closest paint container... of a different color,).

A winding-continuity resistance check wouldn't show any difference, but
an inductance measurement comparison of a known-good transformer
versus one that consistently fries its output transistors might
reveal the problem.

73--

Re: 50 watt Amp IRF 510 Failure

ajparent1/KB1GMX

I can't see.   I literally have no idea.   I am stabbing in the dark.

I gave you information IE, putting in a lot of power and not
getting much out.  Something is wrong.    The fading power if
seen is a cause to hitting the OFF switch.

Wild guess is it acts like it pumping power in to a partial short.
You could have a bad RF cable!  Test that?

Measure the 50 ohm load to make sure its still 50 ohms!

The power divider T2, output transformer T3 could have a short
between turns not between the two wires.  Wild guess.

The output network L4,5,6  could have one extra turn.
Again wild guess.

The capacitors C9, 11, 12, 14 and C18, 19, 20, 21 COULD be
wrong part of value used.

If the transformer is not shorted primary to secondary then the
only thing the ohm meter can tell you is that and not much else.
You can wind the wire on a rusty nail and get that result.

Your could ahve wound it perfectly but the 3turn side in to the
mosfets and not the antenna. Wild guess.

You could have put in any one of the diode backward or wrong
one wrong place we can only guess.

Your trying to measure stuff with a DC instrument and its
apparent only at RF.  Most of the circuit will slow very low
resistances so only gross shorts will show up.

The problem is, its not a common problem!   We are all
at best trying to guess in the dark.

You keep building and likely doing the same thing wrong.
Again that's all guess.

Here at my bench, if thought T3 was backward,  I'd cut it loose and
pull the leads and wind it again.   Its just easier than pondering maybe
IF.  I do it then I'm certain.  To me its a few inches of very cheap wire.
I might even use a bit of black sharpie marker to make one wire
different from the other!  I can imagine if you don't have magnet
wire of the right sizes its a hesitation but not doing that is likely
guessing.  If you are very careful you could desolder the leads
and remove and unwind it and inspect the wires.  Care means
not destroying the board. That can be done, people do it.

If I thought L4, 5 or 6 was an issue I'd carefully de-solder and
count the turns. maybe even take on turn off.   HINT if you count
OUTSIDE turns and get the assembly manual number you have
one too many!

I have a lot of years of prototyping and repairing old gear so
doing it over or fixing without destroying stuff is an acquired
art I guess.

Likely you will have to blast a few more 510s until you figure it out.

Allison
-------------------------------
No direct email, it goes to bit bucket due address harvesting in groups.IO

Re: 10 W linear amplifier: output impedance

ajparent1/KB1GMX

Every time I see that it usually big tube amp with a PI or PI-L
network that when tuned does an exact match. So the question
always seemed rather abstract at best.

Sorta like inquiring on Schrodinger's cats condition, measuring
alters the result.  If you math inclined you can back out the result.

Really and why measure it?

So the numbers are  likely close then but using integer transformers
(full turns only) the translation from drain load impedance to output
impedance in the imperfect world is not ideal.

The closest transformer is 2:3 that translates the 50 ohm load to
22ohms  (that 1.5:1 turns or 2.25:1 impedance translation.)  Its

However the actual source impedance varies with power out and
Drain current. so any measurement is dependent on frequency.
It also varies with bias current.  Finally how the transformer is
would alters it a small amount.

Note that lower source impedance is not a problem as you can
view most solid state amp as power conversion devices.
Termination impedance determined power to load a little
higher is much better than the load being lower resistance.

The only measurement I did was looking into the presumed
drain with a static load to test the transformer.  That load is
a resistor and a capacitor that simulated an actual IRF510
based on actual voltage/current and the data,using  sheet
or measured capacitance. Note all of those fall in a general
area never an exact one.

Why do that?  A VNA does not like a lot of power put into it
and any attenuators mask the data.  Looking backward from
the 50 ohm side and seeing if it hits the desired values
is safe approximation.  That and other design values and
features like the feedback networks the R and C back
to the gate.

Hope that helps.  Its one of those things that is curiosity to
most outside the amplifier designer.

Allison
-------------------------------
No direct email, it goes to bit bucket due address harvesting in groups.IO

Re: 50 watt Amp IRF 510 Failure

George Korper

Allison, I see continuity there. Should I proceed?

That is my question. I am building this second  20 meter amp
to see if I get the same result of sinking power.and if it  is a problem in the way I installed the transformers.
I looked at the schematic and since the 3:2 Transformer is installed first is that a cause of my seeing continuity when the manual says there should not be any continuity?

It has been implied my technique may be the problem so I am being ultra careful.

On Sun, Mar 8, 2020 at 12:18 AM ajparent1/KB1GMX <kb1gmx@...> wrote:
George,

We can only guess, even then we are limited to poorly being devoid of
values, numbers, or test procedure used. Pictures?

The assumption is its a common problem, we all  experience it

What is going on is not common as mine works and does not have
that issue.  Likely how many more do not, most sold?  or we would
have a huge number of people here with a problem

"You should see any continuity here."

I have no clue.

Allison
-------------------------------
No direct email, it goes to bit bucket due address harvesting in groups.IO

Re: 50 watt Amp IRF 510 Failure

ajparent1/KB1GMX

George,

We can only guess, even then we are limited to poorly being devoid of
values, numbers, or test procedure used. Pictures?

The assumption is its a common problem, we all  experience it

What is going on is not common as mine works and does not have
that issue.  Likely how many more do not, most sold?  or we would
have a huge number of people here with a problem

"You should see any continuity here."

I have no clue.

Allison
-------------------------------
No direct email, it goes to bit bucket due address harvesting in groups.IO

Re: 50 watt Amp IRF 510 Failure

George Korper

The sentence: "You should see any continuity here."
What does that mean?

Re: 50 watt Amp IRF 510 Failure

George Korper

Assembly Manual page 17.
I am doing a complete check on  a new board of all my
Transformers as I install them to make sure all my
connections are exactly as Han's intended them.
The bifilar transformer shows continuity from from
the middle pin of Q1 tp the right hand side of C7.
and It says no continuity.

I do not see any obvious errors and if I pull the transformer,
it is not shorted. What is the error?

Re: 50 watt Amp IRF 510 Failure

ajparent1/KB1GMX

That's what I like about groups.io  anything old will be new again after a small wait.

Old info frequently by only days for the email subscriber is often lost information.
Its the nature of the beast.

Allison
-------------------------------
No direct email, it goes to bit bucket due address harvesting in groups.IO

Re: 10 W linear amplifier: output impedance

ajparent1/KB1GMX

Chris look at the multiple posted in the past.
and http://qrp-labs.com/linear.html
Down the page some.

I'd call your numbers low and generally wrong.

That said with the low drive needs (under 50mW, more like 20)
the impedance mismatch at that power level causes no great
issue.

Allison
-------------------------------
No direct email, it goes to bit bucket due address harvesting in groups.IO

Re: Powering the QCX 50 amp via LIPO battery

ajparent1/KB1GMX

Curt,

Checking would be good.  5A at 20V would be 100W and at 70% efficiency minimum
thats 70W out.  I did that as a does it makes sense number.

Typical input is abut 3.5A at 20V or 70-72watts DC and 50 out.'

Big thing for a battery is total watt-hours in the battery . The LiFePo4 is
just lighter for the total watts.

So to be useful you need maybe 140WH battery so a 4S needs to have
cells of 7AH and with 25% duty cycle rate that will give about 8 hours
maybe longer.  That means with 2800mah 18650 cells thats a 4s3p
battery with a battery management board.  OR a packaged battery of
20V 7AH or larger.  FYI the cells need a 2C (or 5A for the battery)
max discharge rate so that usually easy to satisfy.

A lower voltage battery and upconverter has the issue of high peak
current so the discharge rate maybe far higher.  A higher voltages
battery 24V (6S) nominal and buck converter would have a lower peak
current but the converter noise may be an issue and the converter
needs to be 4A peak.

So 6S requires down converter, and the battery has to to not
less than 4A peak.   Raw 6S power is too great and will likely
fry the devices.

Also the 6S pack needs to be at least 140WH capacity or more
for useful run time as your wasting power in the buck converter
(8 to 25%).

If you thinking portable you need to assure the antenna is known good
before going to high power and it can safely operate at that power or
failure is likely.  Also SWR is low (measure and correct) as its pointless
dragging an amp and then frying it.

For field gear aggressively using fuses and bring spares is a good idea.

Allison
-------------------------------