|
All,
To help me post-mortem the Chinese linear,
I was checking the output of the uBITX on various bands. I used a Bird 4410 wattmeter with the 10kW slug ( which can be switched down to 10W full scale ), and my trusty Narda attenuator as a load. Here's what I got
3.615 MHz - 4.2W
7.029 MHz - 4.9W
10.115 MHz - 3.7W
14.030 MHz - 17W <======= HUH?
21.03 MHz - 2.5W ( more like it )
28.35M MHz - 3.9W
That 14MHz reading is a doozy - no WAY is the uBITX making SEVENTEEN WATTS.
My wattmeter ( a recent Ebay purchase ) must be defective! Specifically the slug, where all the RF stuff is.
Let's double check this. I lugged out a scope ( Rigol 1054z ). Plugged it directly into the output of the 30dB Narda attenuator. No termination feature on the scope inputs, and I was too lazy to go get a tee fitting and a 50ohm mini-load. So the readings of the scope would be 6 dB off. Fine, I could live with that, I was interested in the difference, if any, between 20M and the other bands.
40M Hit the mic button - got a reasonable looking sine wave. Little fuzzy.
20M Hit the button - wow, what a mess! Looked like multiple unrelated sine waves at the same time.. Or something. What was it?
Let's look at this in the frequency domain...
Got out the tinySA, plugged it into the output of the Narda. Starting at 80M, I looked for the 3rd harmonic:
80M - 3rd harmonic -32 dB
40M - 3rd harmonic -18 dB
20M - All heck broke loose. Big mess. Spurs all over, in a more or less symetrical pattern around the fundamental. The two biggest ones were at 11.3 MHz ( -10dB ) and 16.55MHz ( -8 dB ). Looked like the unfiltered output of a
mixer.
- Jerry KF6VB
|
Do you think this is the cause of the amp failing, or damage caused by the amp failing and sending some rf back into the radio?
Max KG4PID
On Tuesday, June 8, 2021, 10:27:00 PM CDT, jerry@... <jerry@...> wrote:
All,
To help me post-mortem the Chinese linear,
I was checking the output of the uBITX on various bands.? I used a Bird
4410 wattmeter with the 10kW slug ( which can be switched down to 10W
full scale ), and my trusty Narda attenuator as a load.? Here's what I
got
3.615? MHz - 4.2W
7.029? MHz - 4.9W
10.115 MHz - 3.7W
14.030 MHz - 17W <======= HUH?
21.03? MHz - 2.5W ( more like it )
28.35M MHz - 3.9W
? That 14MHz reading is a doozy - no WAY is the uBITX making SEVENTEEN
WATTS.
My wattmeter ( a recent Ebay purchase ) must be defective!? Specifically
the slug, where all the RF stuff is.
? Let's double check this.? I lugged out a scope ( Rigol 1054z ).?
Plugged it directly into the output of the 30dB Narda attenuator.? No
termination feature on the scope inputs, and I was too lazy to go get a
tee fitting and a 50ohm mini-load.? So the readings of the scope would
be 6 dB off.? Fine, I could live with that, I was interested in the
difference, if any, between 20M and the other bands.
? 40M Hit the mic button - got a reasonable looking sine wave.? Little
fuzzy.
? 20M? Hit the button - wow, what a mess!? Looked like multiple
unrelated sine waves at the same time..? Or something.? What was it?
? Let's look at this in the frequency domain...
Got out the tinySA, plugged it into the output of the Narda.? Starting
at 80M, I looked for the 3rd harmonic:
? 80M - 3rd harmonic -32 dB
? 40M? - 3rd harmonic -18 dB
? 20M? - All heck broke loose.? Big mess.? Spurs all over, in a more or
less symetrical pattern around the fundamental.? The two biggest ones
were at 11.3 MHz ( -10dB ) and 16.55MHz ( -8 dB ).? Looked like the
unfiltered output of a
mixer.
? ? ? ? ? ? ? ? ? ? - Jerry KF6VB
|
Jerry,? I would suspect the measuring instruments. If you have an oscilloscope, then parallel up 20 of 2 watt/50 ohms resistors as a dummy load and measure RF voltage (with bandwidth filter on our scope turned off) across the dummy load (not across the antenna jack).? Once, you are sure of the Rf voltages, then, in parallel to the 50 ohm resistor bundle, add a 2.2K in series with a 50 ohms, such that the 50 ohms is connected to the ground side. Now, measure the RF voltage across the 50 ohms and the square of the ratio of raw to attenuated?voltages is your attenuated power. Crude?by accurate. - f
Do you think this is the cause of the amp failing, or damage caused by the amp failing and sending some rf back into the radio?
Max KG4PID
All,
To help me post-mortem the Chinese linear,
I was checking the output of the uBITX on various bands.? I used a Bird
4410 wattmeter with the 10kW slug ( which can be switched down to 10W
full scale ), and my trusty Narda attenuator as a load.? Here's what I
got
3.615? MHz - 4.2W
7.029? MHz - 4.9W
10.115 MHz - 3.7W
14.030 MHz - 17W <======= HUH?
21.03? MHz - 2.5W ( more like it )
28.35M MHz - 3.9W
? That 14MHz reading is a doozy - no WAY is the uBITX making SEVENTEEN
WATTS.
My wattmeter ( a recent Ebay purchase ) must be defective!? Specifically
the slug, where all the RF stuff is.
? Let's double check this.? I lugged out a scope ( Rigol 1054z ).?
Plugged it directly into the output of the 30dB Narda attenuator.? No
termination feature on the scope inputs, and I was too lazy to go get a
tee fitting and a 50ohm mini-load.? So the readings of the scope would
be 6 dB off.? Fine, I could live with that, I was interested in the
difference, if any, between 20M and the other bands.
? 40M Hit the mic button - got a reasonable looking sine wave.? Little
fuzzy.
? 20M? Hit the button - wow, what a mess!? Looked like multiple
unrelated sine waves at the same time..? Or something.? What was it?
? Let's look at this in the frequency domain...
Got out the tinySA, plugged it into the output of the Narda.? Starting
at 80M, I looked for the 3rd harmonic:
? 80M - 3rd harmonic -32 dB
? 40M? - 3rd harmonic -18 dB
? 20M? - All heck broke loose.? Big mess.? Spurs all over, in a more or
less symetrical pattern around the fundamental.? The two biggest ones
were at 11.3 MHz ( -10dB ) and 16.55MHz ( -8 dB ).? Looked like the
unfiltered output of a
mixer.
? ? ? ? ? ? ? ? ? ? - Jerry KF6VB
|
On 2021-06-08 21:31, Max via groups.io wrote: Do you think this is the cause of the amp failing, or damage caused by
the amp failing and sending some rf back into the radio? *** I don't really know. I think it was on 80M when the amp failed. But I *had* tested it on 20 in that session, and the RF voltage output on 20 was indeed much higher than the other bands. Looks like I need to get into the uBITX and understand it. I'll be reviewing the circuit on paper today. - Jerry Max KG4PID
On Tuesday, June 8, 2021, 10:27:00 PM CDT, jerry@...
<jerry@...> wrote:
All,
To help me post-mortem the Chinese linear,
I was checking the output of the uBITX on various bands. I used a
Bird
4410 wattmeter with the 10kW slug ( which can be switched down to 10W
full scale ), and my trusty Narda attenuator as a load. Here's what I
got
3.615 MHz - 4.2W
7.029 MHz - 4.9W
10.115 MHz - 3.7W
14.030 MHz - 17W <======= HUH?
21.03 MHz - 2.5W ( more like it )
28.35M MHz - 3.9W
That 14MHz reading is a doozy - no WAY is the uBITX making SEVENTEEN
WATTS.
My wattmeter ( a recent Ebay purchase ) must be defective!
Specifically
the slug, where all the RF stuff is.
Let's double check this. I lugged out a scope ( Rigol 1054z ).
Plugged it directly into the output of the 30dB Narda attenuator. No
termination feature on the scope inputs, and I was too lazy to go get
a
tee fitting and a 50ohm mini-load. So the readings of the scope would
be 6 dB off. Fine, I could live with that, I was interested in the
difference, if any, between 20M and the other bands.
40M Hit the mic button - got a reasonable looking sine wave. Little
fuzzy.
20M Hit the button - wow, what a mess! Looked like multiple
unrelated sine waves at the same time.. Or something. What was it?
Let's look at this in the frequency domain...
Got out the tinySA, plugged it into the output of the Narda. Starting
at 80M, I looked for the 3rd harmonic:
80M - 3rd harmonic -32 dB
40M - 3rd harmonic -18 dB
20M - All heck broke loose. Big mess. Spurs all over, in a more
or
less symetrical pattern around the fundamental. The two biggest ones
were at 11.3 MHz ( -10dB ) and 16.55MHz ( -8 dB ). Looked like the
unfiltered output of a
mixer.
- Jerry KF6VB
Links:
------
[1] /g/BITX20/message/89053
[2] /mt/83413015/243852
[3] /g/BITX20/post
[4] /g/BITX20/editsub/243852
[5] /g/BITX20/leave/10189903/243852/952924773/xyzzy
|
Ashhar,
I would expect the effect - of the scope not being terminated - to
be minimal, because there was a 30dB Narda attenuator between the uBITX
and the scope. The Narda is rated for microwave, and is flat as a board through
HF & VHF. And the difference at the input - what the uBITX sees - between
terminated and not terminated ( at the output of the Narda ) - is minuscule. Because it's a 30dB attenuator - divides the power by 1000. So 99.9% of the power put into that attenuator is dissipated by the attenuator itself.
The idea was not to use the scope to *measure* the uBITX output in any absolute sense, but rather to compare it in a qualitative sense to the output of the uBITX
on other bands.
The tinySA verified massive mixing product spurs on 20. And it *does* have an internal termination.
- Jerry KF6VB
toggle quoted message
Show quoted text
On 2021-06-08 21:45, Ashhar Farhan wrote: Jerry,
I would suspect the measuring instruments. If you have an
oscilloscope, then parallel up 20 of 2 watt/50 ohms resistors as a
dummy load and measure RF voltage (with bandwidth filter on our scope
turned off) across the dummy load (not across the antenna jack).
Once, you are sure of the Rf voltages, then, in parallel to the 50 ohm
resistor bundle, add a 2.2K in series with a 50 ohms, such that the 50
ohms is connected to the ground side. Now, measure the RF voltage
across the 50 ohms and the square of the ratio of raw to attenuated
voltages is your attenuated power. Crude by accurate.
- f
On Wed, Jun 9, 2021 at 10:02 AM Max via groups.io [1]
<kg4pid@...> wrote:
Do you think this is the cause of the amp failing, or damage caused
by the amp failing and sending some rf back into the radio?
Max KG4PID
On Tuesday, June 8, 2021, 10:27:00 PM CDT, jerry@...
<jerry@...> wrote:
All,
To help me postmortem the Chinese linear,
I was checking the output of the uBITX on various bands. I used a
Bird
4410 wattmeter with the 10kW slug ( which can be switched down to
10W
full scale ), and my trusty Narda attenuator as a load. Here's what
I
got
3.615 MHz - 4.2W
7.029 MHz - 4.9W
10.115 MHz - 3.7W
14.030 MHz - 17W <======= HUH?
21.03 MHz - 2.5W ( more like it )
28.35M MHz - 3.9W
That 14MHz reading is a doozy - no WAY is the uBITX making
SEVENTEEN
WATTS.
My wattmeter ( a recent Ebay purchase ) must be defective!
Specifically
the slug, where all the RF stuff is.
Let's double check this. I lugged out a scope ( Rigol 1054z ).
Plugged it directly into the output of the 30dB Narda attenuator.
No
termination feature on the scope inputs, and I was too lazy to go
get a
tee fitting and a 50ohm mini-load. So the readings of the scope
would
be 6 dB off. Fine, I could live with that, I was interested in the
difference, if any, between 20M and the other bands.
40M Hit the mic button - got a reasonable looking sine wave.
Little
fuzzy.
20M Hit the button - wow, what a mess! Looked like multiple
unrelated sine waves at the same time.. Or something. What was it?
Let's look at this in the frequency domain...
Got out the tinySA, plugged it into the output of the Narda.
Starting
at 80M, I looked for the 3rd harmonic:
80M - 3rd harmonic -32 dB
40M - 3rd harmonic -18 dB
20M - All heck broke loose. Big mess. Spurs all over, in a more
or
less symetrical pattern around the fundamental. The two biggest
ones
were at 11.3 MHz ( -10dB ) and 16.55MHz ( -8 dB ). Looked like the
unfiltered output of a
mixer.
- Jerry KF6VB Links:
------
[1]
[2] /g/BITX20/message/89054
[3] /mt/83413015/243852
[4] /g/BITX20/post
[5] /g/BITX20/editsub/243852
[6] /g/BITX20/leave/10189903/243852/952924773/xyzzy
|
The tinysa does not have shielding needed for a decent, spur free reading, at 40 watts of power it will probably read spurs even without connecting it up.?
Ashhar,
? ? I would expect the effect - of the scope not being terminated - to
be minimal, because there was a 30dB Narda attenuator between the uBITX
and the scope.? The Narda is rated for microwave, and is flat as a board
through
HF & VHF.? And the difference at the input - what the uBITX sees -
between
terminated and not terminated ( at the output of the Narda ) - is
minuscule.? Because it's a 30dB attenuator - divides the power by 1000.?
So 99.9% of the power put into that attenuator is dissipated by the
attenuator itself.
? ?The idea was not to use the scope to *measure* the uBITX output in any
absolute sense, but rather to compare it in a qualitative sense to the
output of the uBITX
on other bands.
? ?The tinySA verified massive mixing product spurs on 20.? And it *does*
have an internal termination.
? ? ? ? ? ? ? ? ? ? ? ? - Jerry KF6VB
On 2021-06-08 21:45, Ashhar Farhan wrote:
> Jerry,
> I would suspect the measuring instruments. If you have an
> oscilloscope, then parallel up 20 of 2 watt/50 ohms resistors as a
> dummy load and measure RF voltage (with bandwidth filter on our scope
> turned off) across the dummy load (not across the antenna jack).
> Once, you are sure of the Rf voltages, then, in parallel to the 50 ohm
> resistor bundle, add a 2.2K in series with a 50 ohms, such that the 50
> ohms is connected to the ground side. Now, measure the RF voltage
> across the 50 ohms and the square of the ratio of raw to attenuated
> voltages is your attenuated power. Crude by accurate.
> - f
>
> On Wed, Jun 9, 2021 at 10:02 AM Max via [1]
> <kg4pid=[email protected]> wrote:
>
>> Do you think this is the cause of the amp failing, or damage caused
>> by the amp failing and sending some rf back into the radio?
>>
>> Max KG4PID
>>
>> On Tuesday, June 8, 2021, 10:27:00 PM CDT, jerry@...
>> <jerry@...> wrote:
>>
>> All,
>>
>> To help me postmortem the Chinese linear,
>>
>> I was checking the output of the uBITX on various bands.? I used a
>> Bird
>>
>> 4410 wattmeter with the 10kW slug ( which can be switched down to
>> 10W
>>
>> full scale ), and my trusty Narda attenuator as a load.? Here's what
>> I
>>
>> got
>>
>> 3.615? MHz - 4.2W
>>
>> 7.029? MHz - 4.9W
>>
>> 10.115 MHz - 3.7W
>>
>> 14.030 MHz - 17W <======= HUH?
>>
>> 21.03? MHz - 2.5W ( more like it )
>>
>> 28.35M MHz - 3.9W
>>
>> That 14MHz reading is a doozy - no WAY is the uBITX making
>> SEVENTEEN
>>
>> WATTS.
>>
>> My wattmeter ( a recent Ebay purchase ) must be defective!
>> Specifically
>>
>> the slug, where all the RF stuff is.
>>
>> Let's double check this.? I lugged out a scope ( Rigol 1054z ).
>>
>> Plugged it directly into the output of the 30dB Narda attenuator.
>> No
>>
>> termination feature on the scope inputs, and I was too lazy to go
>> get a
>>
>> tee fitting and a 50ohm mini-load.? So the readings of the scope
>> would
>>
>> be 6 dB off.? Fine, I could live with that, I was interested in the
>>
>> difference, if any, between 20M and the other bands.
>>
>> 40M Hit the mic button - got a reasonable looking sine wave.
>> Little
>>
>> fuzzy.
>>
>> 20M? Hit the button - wow, what a mess!? Looked like multiple
>>
>> unrelated sine waves at the same time..? Or something.? What was it?
>>
>> Let's look at this in the frequency domain...
>>
>> Got out the tinySA, plugged it into the output of the Narda.
>> Starting
>>
>> at 80M, I looked for the 3rd harmonic:
>>
>> 80M - 3rd harmonic -32 dB
>>
>> 40M? - 3rd harmonic -18 dB
>>
>> 20M? - All heck broke loose.? Big mess.? Spurs all over, in a more
>> or
>>
>> less symetrical pattern around the fundamental.? The two biggest
>> ones
>>
>> were at 11.3 MHz ( -10dB ) and 16.55MHz ( -8 dB ).? Looked like the
>>
>> unfiltered output of a
>>
>> mixer.
>>
>> - Jerry KF6VB
>
>
>
> Links:
> ------
> [1]
> [2] /g/BITX20/message/89054
> [3] /mt/83413015/243852
> [4] /g/BITX20/post
> [5] /g/BITX20/editsub/243852
> [6] /g/BITX20/leave/10189903/243852/952924773/xyzzy
|
Hi Farhhan,
I'll hook up my other spectrum analyzer today. It's a Texscan
VSM-5D. Only 4MHz-450MHz, but that's plenty for this.
HOWEVER, this test was without any amp. Just the uBITX feeding the
Narda attenuator. And the scope & the tinySA agree - lots of mixing products. Only on 20M.
Reading the block diagram, I am at a loss as to how such a problem could
occur. The two biggest spurs are at the fundamental plus about 2.5Mhz,
and also MINUS 2.5Mhz.
Specific to 20M would only be the LPF. And what failure in the LPF could cause such a cacophony of mixing spurs?
I will also check the outputs of the si5351a - I am using the TSW Teensy board, will try popping your display/vfo board back in.
- Jerry KF6VB
toggle quoted message
Show quoted text
On 2021-06-09 08:51, Ashhar Farhan wrote: The tinysa does not have shielding needed for a decent, spur free
reading, at 40 watts of power it will probably read spurs even without
connecting it up.
On Wed 9 Jun, 2021, 6:41 PM jerry@..., <jerry@...> wrote:
Ashhar,
I would expect the effect - of the scope not being terminated -
to
be minimal, because there was a 30dB Narda attenuator between the
uBITX
and the scope. The Narda is rated for microwave, and is flat as a
board
through
HF & VHF. And the difference at the input - what the uBITX sees -
between
terminated and not terminated ( at the output of the Narda ) - is
minuscule. Because it's a 30dB attenuator - divides the power by
1000.
So 99.9% of the power put into that attenuator is dissipated by the
attenuator itself.
The idea was not to use the scope to *measure* the uBITX output
in any
absolute sense, but rather to compare it in a qualitative sense to
the
output of the uBITX
on other bands.
The tinySA verified massive mixing product spurs on 20. And it
*does*
have an internal termination.
- Jerry KF6VB
On 2021-06-08 21:45, Ashhar Farhan wrote:
Jerry,
I would suspect the measuring instruments. If you have an
oscilloscope, then parallel up 20 of 2 watt/50 ohms resistors as a
dummy load and measure RF voltage (with bandwidth filter on our scope
turned off) across the dummy load (not across the antenna jack).
Once, you are sure of the Rf voltages, then, in parallel to the 50 ohm
resistor bundle, add a 2.2K in series with a 50 ohms, such that the 50
ohms is connected to the ground side. Now, measure the RF voltage
across the 50 ohms and the square of the ratio of raw to attenuated
voltages is your attenuated power. Crude by accurate.
- f
On Wed, Jun 9, 2021 at 10:02 AM Max via groups.io [1] [1]
<kg4pid@...> wrote:
Do you think this is the cause of the amp failing, or damage caused
by the amp failing and sending some rf back into the radio?
Max KG4PID
On Tuesday, June 8, 2021, 10:27:00 PM CDT, jerry@...
<jerry@...> wrote:
All,
To help me postmortem the Chinese linear,
I was checking the output of the uBITX on various bands. I used
a
Bird
4410 wattmeter with the 10kW slug ( which can be switched down to
10W
full scale ), and my trusty Narda attenuator as a load. Here's
what
I
got
3.615 MHz - 4.2W
7.029 MHz - 4.9W
10.115 MHz - 3.7W
14.030 MHz - 17W <======= HUH?
21.03 MHz - 2.5W ( more like it )
28.35M MHz - 3.9W
That 14MHz reading is a doozy - no WAY is the uBITX making
SEVENTEEN
WATTS.
My wattmeter ( a recent Ebay purchase ) must be defective!
Specifically
the slug, where all the RF stuff is.
Let's double check this. I lugged out a scope ( Rigol 1054z ).
Plugged it directly into the output of the 30dB Narda attenuator.
No
termination feature on the scope inputs, and I was too lazy to go
get a
tee fitting and a 50ohm mini-load. So the readings of the scope
would
be 6 dB off. Fine, I could live with that, I was interested in
the
difference, if any, between 20M and the other bands.
40M Hit the mic button - got a reasonable looking sine wave.
Little
fuzzy.
20M Hit the button - wow, what a mess! Looked like multiple
unrelated sine waves at the same time.. Or something. What was
it?
Let's look at this in the frequency domain...
Got out the tinySA, plugged it into the output of the Narda.
Starting
at 80M, I looked for the 3rd harmonic:
80M - 3rd harmonic -32 dB
40M - 3rd harmonic -18 dB
20M - All heck broke loose. Big mess. Spurs all over, in a
more
or
less symetrical pattern around the fundamental. The two biggest
ones
were at 11.3 MHz ( -10dB ) and 16.55MHz ( -8 dB ). Looked like
the
unfiltered output of a
mixer.
- Jerry KF6VB Links:
------
[1]
[2] /g/BITX20/message/89054
[3] /mt/83413015/243852
[4] /g/BITX20/post
[5] /g/BITX20/editsub/243852
[6]
/g/BITX20/leave/10189903/243852/952924773/xyzzy Links:
------
[1]
[2] /g/BITX20/message/89060
[3] /mt/83413015/243852
[4] /g/BITX20/post
[5] /g/BITX20/editsub/243852
[6] /g/BITX20/leave/10189903/243852/952924773/xyzzy
|
Which version of ubitx is this? The versions prior to v5 needed an extra lpf between the first mixer and the if amp?
Hi Farhhan,
I'll hook up my other spectrum analyzer today.? It's a Texscan
VSM-5D.? Only 4MHz-450MHz, but that's plenty for this.
HOWEVER, this test was without any amp.? Just the uBITX feeding the
Narda attenuator.? And the scope & the tinySA agree - lots of mixing
products.? Only on 20M.
Reading the block diagram, I am at a loss as to how such a problem could
occur.? The two biggest spurs are at the fundamental plus about 2.5Mhz,
and also MINUS 2.5Mhz.
Specific to 20M would only be the LPF.? And what failure in the LPF
could cause such a cacophony of mixing spurs?
I will also check the outputs of the si5351a - I am using the TSW Teensy
board, will try popping your display/vfo board back in.
? ? ? ? ? ? ? ? ? ? ? ? - Jerry KF6VB
On 2021-06-09 08:51, Ashhar Farhan wrote:
> The tinysa does not have shielding needed for a decent, spur free
> reading, at 40 watts of power it will probably read spurs even without
> connecting it up.
>
> On Wed 9 Jun, 2021, 6:41 PM jerry@..., <jerry@...> wrote:
>
>> Ashhar,
>>
>> I would expect the effect - of the scope not being terminated -
>> to
>> be minimal, because there was a 30dB Narda attenuator between the
>> uBITX
>> and the scope.? The Narda is rated for microwave, and is flat as a
>> board
>> through
>> HF & VHF.? And the difference at the input - what the uBITX sees -
>> between
>> terminated and not terminated ( at the output of the Narda ) - is
>> minuscule.? Because it's a 30dB attenuator - divides the power by
>> 1000.
>> So 99.9% of the power put into that attenuator is dissipated by the
>> attenuator itself.
>>
>> The idea was not to use the scope to *measure* the uBITX output
>> in any
>> absolute sense, but rather to compare it in a qualitative sense to
>> the
>> output of the uBITX
>> on other bands.
>>
>> The tinySA verified massive mixing product spurs on 20.? And it
>> *does*
>> have an internal termination.
>>
>> - Jerry KF6VB
>>
>> On 2021-06-08 21:45, Ashhar Farhan wrote:
>>> Jerry,
>>> I would suspect the measuring instruments. If you have an
>>> oscilloscope, then parallel up 20 of 2 watt/50 ohms resistors as a
>>> dummy load and measure RF voltage (with bandwidth filter on our
>> scope
>>> turned off) across the dummy load (not across the antenna jack).
>>> Once, you are sure of the Rf voltages, then, in parallel to the 50
>> ohm
>>> resistor bundle, add a 2.2K in series with a 50 ohms, such that
>> the 50
>>> ohms is connected to the ground side. Now, measure the RF voltage
>>> across the 50 ohms and the square of the ratio of raw to
>> attenuated
>>> voltages is your attenuated power. Crude by accurate.
>>> - f
>>>
>>> On Wed, Jun 9, 2021 at 10:02 AM Max via [1] [1]
>>> <kg4pid=[email protected]> wrote:
>>>
>>>> Do you think this is the cause of the amp failing, or damage
>> caused
>>>> by the amp failing and sending some rf back into the radio?
>>>>
>>>> Max KG4PID
>>>>
>>>> On Tuesday, June 8, 2021, 10:27:00 PM CDT, jerry@...
>>>> <jerry@...> wrote:
>>>>
>>>> All,
>>>>
>>>> To help me postmortem the Chinese linear,
>>>>
>>>> I was checking the output of the uBITX on various bands.? I used
>> a
>>>> Bird
>>>>
>>>> 4410 wattmeter with the 10kW slug ( which can be switched down to
>>>> 10W
>>>>
>>>> full scale ), and my trusty Narda attenuator as a load.? Here's
>> what
>>>> I
>>>>
>>>> got
>>>>
>>>> 3.615? MHz - 4.2W
>>>>
>>>> 7.029? MHz - 4.9W
>>>>
>>>> 10.115 MHz - 3.7W
>>>>
>>>> 14.030 MHz - 17W <======= HUH?
>>>>
>>>> 21.03? MHz - 2.5W ( more like it )
>>>>
>>>> 28.35M MHz - 3.9W
>>>>
>>>> That 14MHz reading is a doozy - no WAY is the uBITX making
>>>> SEVENTEEN
>>>>
>>>> WATTS.
>>>>
>>>> My wattmeter ( a recent Ebay purchase ) must be defective!
>>>> Specifically
>>>>
>>>> the slug, where all the RF stuff is.
>>>>
>>>> Let's double check this.? I lugged out a scope ( Rigol 1054z ).
>>>>
>>>> Plugged it directly into the output of the 30dB Narda attenuator.
>>>> No
>>>>
>>>> termination feature on the scope inputs, and I was too lazy to go
>>>> get a
>>>>
>>>> tee fitting and a 50ohm mini-load.? So the readings of the scope
>>>> would
>>>>
>>>> be 6 dB off.? Fine, I could live with that, I was interested in
>> the
>>>>
>>>> difference, if any, between 20M and the other bands.
>>>>
>>>> 40M Hit the mic button - got a reasonable looking sine wave.
>>>> Little
>>>>
>>>> fuzzy.
>>>>
>>>> 20M? Hit the button - wow, what a mess!? Looked like multiple
>>>>
>>>> unrelated sine waves at the same time..? Or something.? What was
>> it?
>>>>
>>>> Let's look at this in the frequency domain...
>>>>
>>>> Got out the tinySA, plugged it into the output of the Narda.
>>>> Starting
>>>>
>>>> at 80M, I looked for the 3rd harmonic:
>>>>
>>>> 80M - 3rd harmonic -32 dB
>>>>
>>>> 40M? - 3rd harmonic -18 dB
>>>>
>>>> 20M? - All heck broke loose.? Big mess.? Spurs all over, in a
>> more
>>>> or
>>>>
>>>> less symetrical pattern around the fundamental.? The two biggest
>>>> ones
>>>>
>>>> were at 11.3 MHz ( -10dB ) and 16.55MHz ( -8 dB ).? Looked like
>> the
>>>>
>>>> unfiltered output of a
>>>>
>>>> mixer.
>>>>
>>>> - Jerry KF6VB
>>>
>>>
>>>
>>> Links:
>>> ------
>>> [1]
>>> [2] /g/BITX20/message/89054
>>> [3] /mt/83413015/243852
>>> [4] /g/BITX20/post
>>> [5] /g/BITX20/editsub/243852
>>> [6]
>> /g/BITX20/leave/10189903/243852/952924773/xyzzy
>
>
>
> Links:
> ------
> [1]
> [2] /g/BITX20/message/89060
> [3] /mt/83413015/243852
> [4] /g/BITX20/post
> [5] /g/BITX20/editsub/243852
> [6] /g/BITX20/leave/10189903/243852/952924773/xyzzy
|
On 2021-06-09 19:06, Ashhar Farhan wrote: Which version of ubitx is this? The versions prior to v5 needed an
extra lpf between the first mixer and the if amp V6 - Jerry On Wed 9 Jun, 2021, 10:31 PM jerry@..., <jerry@...> wrote:
Hi Farhhan,
I'll hook up my other spectrum analyzer today. It's a Texscan
VSM-5D. Only 4MHz-450MHz, but that's plenty for this.
HOWEVER, this test was without any amp. Just the uBITX feeding the
Narda attenuator. And the scope & the tinySA agree - lots of mixing
products. Only on 20M.
Reading the block diagram, I am at a loss as to how such a problem
could
occur. The two biggest spurs are at the fundamental plus about
2.5Mhz,
and also MINUS 2.5Mhz.
Specific to 20M would only be the LPF. And what failure in the LPF
could cause such a cacophony of mixing spurs?
I will also check the outputs of the si5351a - I am using the TSW
Teensy
board, will try popping your display/vfo board back in.
- Jerry KF6VB
On 2021-06-09 08:51, Ashhar Farhan wrote:
The tinysa does not have shielding needed for a decent, spur free
reading, at 40 watts of power it will probably read spurs even without
connecting it up.
On Wed 9 Jun, 2021, 6:41 PM jerry@..., <jerry@...> wrote:
Ashhar,
I would expect the effect - of the scope not being terminated -
to
be minimal, because there was a 30dB Narda attenuator between the
uBITX
and the scope. The Narda is rated for microwave, and is flat as a
board
through
HF & VHF. And the difference at the input - what the uBITX sees
-
between
terminated and not terminated ( at the output of the Narda ) - is
minuscule. Because it's a 30dB attenuator - divides the power by
1000.
So 99.9% of the power put into that attenuator is dissipated by
the
attenuator itself.
The idea was not to use the scope to *measure* the uBITX output
in any
absolute sense, but rather to compare it in a qualitative sense
to
the
output of the uBITX
on other bands.
The tinySA verified massive mixing product spurs on 20. And it
*does*
have an internal termination.
- Jerry KF6VB
On 2021-06-08 21:45, Ashhar Farhan wrote:
Jerry,
I would suspect the measuring instruments. If you have an
oscilloscope, then parallel up 20 of 2 watt/50 ohms resistors as
a
dummy load and measure RF voltage (with bandwidth filter on our scope
turned off) across the dummy load (not across the antenna jack).
Once, you are sure of the Rf voltages, then, in parallel to the
50
ohm
resistor bundle, add a 2.2K in series with a 50 ohms, such that the 50
ohms is connected to the ground side. Now, measure the RF
voltage
across the 50 ohms and the square of the ratio of raw to attenuated
voltages is your attenuated power. Crude by accurate.
- f
On Wed, Jun 9, 2021 at 10:02 AM Max via groups.io [1] [1] [1]
<kg4pid@...> wrote:
Do you think this is the cause of the amp failing, or damage caused
by the amp failing and sending some rf back into the radio?
Max KG4PID
On Tuesday, June 8, 2021, 10:27:00 PM CDT, jerry@...
<jerry@...> wrote:
All,
To help me postmortem the Chinese linear,
I was checking the output of the uBITX on various bands. I
used
a
Bird
4410 wattmeter with the 10kW slug ( which can be switched down
to
10W
full scale ), and my trusty Narda attenuator as a load. Here's
what
I
got
3.615 MHz - 4.2W
7.029 MHz - 4.9W
10.115 MHz - 3.7W
14.030 MHz - 17W <======= HUH?
21.03 MHz - 2.5W ( more like it )
28.35M MHz - 3.9W
That 14MHz reading is a doozy - no WAY is the uBITX making
SEVENTEEN
WATTS.
My wattmeter ( a recent Ebay purchase ) must be defective!
Specifically
the slug, where all the RF stuff is.
Let's double check this. I lugged out a scope ( Rigol 1054z ).
Plugged it directly into the output of the 30dB Narda
attenuator.
No
termination feature on the scope inputs, and I was too lazy to
go
get a
tee fitting and a 50ohm mini-load. So the readings of the
scope
would
be 6 dB off. Fine, I could live with that, I was interested in
the
difference, if any, between 20M and the other bands.
40M Hit the mic button - got a reasonable looking sine wave.
Little
fuzzy.
20M Hit the button - wow, what a mess! Looked like multiple
unrelated sine waves at the same time.. Or something. What
was
it?
Let's look at this in the frequency domain...
Got out the tinySA, plugged it into the output of the Narda.
Starting
at 80M, I looked for the 3rd harmonic:
80M - 3rd harmonic -32 dB
40M - 3rd harmonic -18 dB
20M - All heck broke loose. Big mess. Spurs all over, in a
more
or
less symetrical pattern around the fundamental. The two
biggest
ones
were at 11.3 MHz ( -10dB ) and 16.55MHz ( -8 dB ). Looked like
the
unfiltered output of a
mixer.
- Jerry KF6VB Links:
------
[1]
[2] /g/BITX20/message/89054
[3] /mt/83413015/243852
[4] /g/BITX20/post
[5] /g/BITX20/editsub/243852
[6]
/g/BITX20/leave/10189903/243852/952924773/xyzzy
Links:
------
[1]
[2] /g/BITX20/message/89060
[3] /mt/83413015/243852
[4] /g/BITX20/post
[5] /g/BITX20/editsub/243852
[6]
/g/BITX20/leave/10189903/243852/952924773/xyzzy Links:
------
[1]
[2] /g/BITX20/message/89073
[3] /mt/83413015/243852
[4] /g/BITX20/post
[5] /g/BITX20/editsub/243852
[6] /g/BITX20/leave/10189903/243852/952924773/xyzzy
|
From what I have seen with my V6, you will see the 2 spurs with your good spectrum analyzer also. I installed a HPF since mine showed up on 21Mhz and up
John
|
n 2021-06-09 19:40, John Cunliffe W7ZQ wrote: From what I have seen with my V6, you will see the 2 spurs with >your good spectrum analyzer also. I installed a HPF since mine showed >up on
21Mhz and up
Was it this bad? See attached picture of tinySA trace.. I just swapped out the TSW Raduino and reinstalled the HFSignals one. No difference. I thought maybe since the si5351 was on that card then... Note that the closest spurs are only 8 & 10 dB below the fundamental. That means that if I'm transmitting 5W on 14 MHz, O I'm also transmitting a full half a watt on 16MHz...which isn't a ham band. Neither is 12 MHz. Share some details about your HPF? Did you put it on the output, or insert it into the circuit? How many poles? For my radio to be legal, a 20M HPF would have to give over 30 dB of attenuation at 2 MHz away from the fundamental. - Jerry John
Links:
------
[1] /g/BITX20/message/89075
[2] /mt/83413015/243852
[3] /g/BITX20/post
[4] /g/BITX20/editsub/243852
[5] /g/BITX20/leave/10189903/243852/952924773/xyzzy
|
Jerry,
To help me understand better what is going on, can you confirm that the testing is using CW?? Also, what, if any modifications have you done to the v6.
Thank you,
73
Evan
AC9TU
|
Jerry,
Can this be some form of self-oscillation? It is so bad, this looks like single-tone FM!!
Rafael Pinto
|
Hi Evan,
Yes, CW. Using the mic button to key, with the speed set to 1WPM. Have not modified the V6. The test setup:
v6 -> Narda 30dB attenuator -> tinySA.
- Jerry
toggle quoted message
Show quoted text
|
Rafael,
I was thinking exactly this last night. Self oscillation of the final stage around 2MHz? Mixing with the outgoing CW?
If so, maybe it would oscillate also on SSB mode? I'll try that
today.
- Jerry KF6VB
toggle quoted message
Show quoted text
|
I am looking a the schematic of the transmit chain. I possibly need to check all the bypass caps. C86, C92 etc. Also the
negative feedback caps, C261 & C262. And the values of the
negative feedback resistors. Bypass caps can be checked by scoping the hot end for RF.
I used to do this stuff for a living, but that was 40 years ago.
Hopefully it's like riding a bicycle :). If it is an oscillation,
those are hard to find because everything oscillates at once...
- Jerry KF6VB
toggle quoted message
Show quoted text
|
My Ubitx? ( From India) has this problem.? Basically its joined the queue? in the corner of the shack of stuff needing attention. I'll get round to it some day but in the meantime my old K1 is doing the QRP job.
I suspect lack of QC at the assembly point .
Larry G4GZG.
|
Jerry,
Maybe a signal comparison with an oscilloscope starting with TP3 to TP7.? Other than the half-wave aspect of TP5 and TP7 it should be just an amplified signal.
Just a thought.
73
Evan
AC9TU
|
Here's a tinySA of the output on 40M. Much nicer than 20, but still not good. 3rd harmonic is only 23dB below the fundamental. Other bands ( besides 20 ) are similar.
- Jerry KF6VB
|
A quick inspection of the v6 motherboard...
C261 & C262....
....Seem to be missing.
These are the negative feedback caps for the final transistors,
Q94 & Q95. There are surface mount lands for them, but no parts
installed.
Their associated resistors, R261 & R262, *are* installed.
Exact spec for these capacitors? Voltage rating, SMD size etc?
Not going to be easy to install these - they're almost right under the final transistors. Tweezers and smd rework air gun...
- Jerry KF6VB
toggle quoted message
Show quoted text
On 2021-06-10 07:32, jerry@... wrote: Here's a tinySA of the output on 40M. Much nicer than 20, but still
not good. 3rd harmonic is only 23dB below the fundamental. Other
bands ( besides 20 ) are similar.
- Jerry KF6VB
|
Rafael Pinto [PU1OWL]
