Date   
Re: No TX #ubitx

Ted
 

The bias controls do make the expected adjustments in current for the finals.

Interestingly enough, I was repeating some voltage readings yesterday and when I put my VOM lead into Pin 1 of Q90, the bottom of my hand - which was touching the IRF510 heat sinks- started to get warm.  A glance toward the wattmeter told me there was RF out!

The TX output quit when I removed the lead and returned more or less when touching the Q90 pin again. After touching the side points for 2cm around in all directions with nothing to show for it, I went and threw in, a 2N2222 as an upgrade, taking care to bend the gate backward betwixt the other two leads so that the E., B., and C. were on the circuit board pads they need to be on.  In firing the rig on, the '2222 gets nice and toasty, as does the pre driver pair (as they did before).

So, did that fix the problem? Actually, no. Still no RF out, now.  I put a 2N3904 back in and still nothing, though, being as the pre drivers are warm as well, I'd suspect the problem comes after them. An RF "sniffer" loop shows that RF fields diminish quickly, downstream of Q90 toward the rest of the transmitter power stages. 

I spent the time believing in a Q90 stage solution because of the sporadic full-strength output while taking measurements. It defies explanation how simply making metallic contact can restore seemingly full output, in a way that a re-solder and/or Q90 replacement didn't fix.

I'm sure that reading this report makes no sense. Well, neither did it make sense being there at the time so we're in the same place on that count.

Diving back in after the day job is done with.



-Ted
K3RTA


On Sat, Feb 16, 2019 at 6:00, Raj vu2zap
<rajendrakumargg@...> wrote:
Ted,

does the bias control pots change the idle current ? if it does then finals are fine and final circuit is fine!

Raj

At 16/02/2019, you wrote:
>On Sun, Feb 10, 2019 at 11:11 PM, Raj vu2zap wrote:
>
>
>Your finals are blown. Check if you have TX 5V at the output of regulator, if yes
>then replace IRF510's.
>--------------------------
>
>Nope, new finals made no difference, surprisingly enough.  Have put this aside for a few days to get a fresh start next time diving into this. It's interesting how such a seemingly simple thing gets complicated. 
>
>See my update above for the latest readings, if you like.  I'd love to see this back on line before my V5 package shows up next week :)
>
>
>Regards,
>
>Ted
>K3RTA




Re: No TX #ubitx

Adrian Chadd
 

Hi,

did you test the existing transistor you removed to see if it was actually a functioning transistor still? Q90 doesn't look like it should toasty?


-a

Re: #bitx40 #bitx40

Scot McMath
 

I'm in AZ, how much do you want for 1 of your Bitx 40's? I can pay shipping costs.

Re: load ver.4 original firmware into ubitx ver.3

Vic WA4THR
 

The v4 code has a different calibration scheme, which is actually pretty easy if you have a second receiver. The code is posted at: https://github.com/afarhan/ubitx4

You can get the needed Arduino IDE compiler for free at: https://www.arduino.cc/en/main/software

Using the Arduino IDE program, you can read the ubitx_v4.3_code.ino code, compile it, and load it into the Raduino with a USB cable.

While I now understand this, I'll admit the process was not so obvious when I started. Another "geezer" (AA4PG) posted some very clear instructions here: http://radman.no-ip.ca:96/wordpress/?p=2747

=Vic=

Re: No TX #ubitx

Ted
 

^ That is, I touched points for about 2cm around Q90 with a soldering pencil, checking for cold joints. Etc.

Re: #bitx40 #bitx40

Ted
 

I could be tempted to own one, if you'd ship to the Mid-Atlantic.


-Ted
K3RTA

Re: No TX #ubitx

jim
 

And the input Z of your vom is? (Ohms per Volt) ..and the reading was  ? ...Just possible the vom was providing enough bias to turn on Q90 due to its loading

Jim

On Monday, February 18, 2019, 9:54:14 AM PST, Ted via Groups.Io <k3rta@...> wrote:


The bias controls do make the expected adjustments in current for the finals.

Interestingly enough, I was repeating some voltage readings yesterday and when I put my VOM lead into Pin 1 of Q90, the bottom of my hand - which was touching the IRF510 heat sinks- started to get warm.  A glance toward the wattmeter told me there was RF out!

The TX output quit when I removed the lead and returned more or less when touching the Q90 pin again. After touching the side points for 2cm around in all directions with nothing to show for it, I went and threw in, a 2N2222 as an upgrade, taking care to bend the gate backward betwixt the other two leads so that the E., B., and C. were on the circuit board pads they need to be on.  In firing the rig on, the '2222 gets nice and toasty, as does the pre driver pair (as they did before).

So, did that fix the problem? Actually, no. Still no RF out, now.  I put a 2N3904 back in and still nothing, though, being as the pre drivers are warm as well, I'd suspect the problem comes after them. An RF "sniffer" loop shows that RF fields diminish quickly, downstream of Q90 toward the rest of the transmitter power stages. 

I spent the time believing in a Q90 stage solution because of the sporadic full-strength output while taking measurements. It defies explanation how simply making metallic contact can restore seemingly full output, in a way that a re-solder and/or Q90 replacement didn't fix.

I'm sure that reading this report makes no sense. Well, neither did it make sense being there at the time so we're in the same place on that count.

Diving back in after the day job is done with.



-Ted
K3RTA


On Sat, Feb 16, 2019 at 6:00, Raj vu2zap
<rajendrakumargg@...> wrote:
Ted,

does the bias control pots change the idle current ? if it does then finals are fine and final circuit is fine!

Raj

At 16/02/2019, you wrote:
>On Sun, Feb 10, 2019 at 11:11 PM, Raj vu2zap wrote:
>
>
>Your finals are blown. Check if you have TX 5V at the output of regulator, if yes
>then replace IRF510's.
>--------------------------
>
>Nope, new finals made no difference, surprisingly enough.  Have put this aside for a few days to get a fresh start next time diving into this. It's interesting how such a seemingly simple thing gets complicated. 
>
>See my update above for the latest readings, if you like.  I'd love to see this back on line before my V5 package shows up next week :)
>
>
>Regards,
>
>Ted
>K3RTA




Re: No TX #ubitx

 

Ted,

if touching Q90 make the power amp oscillate then probably the whole chain is ok. This is quite normal for a
high gain PA.

now check L1/2/3/4 for continuity, there could be a bad solder on one of the toroids.

There is a spreadsheet in a folder called vu2zap in the file section with static voltages of uBitxv4 if you want
to check.

Raj


At 18-02-19, you wrote:
The bias controls do make the expected adjustments in current for the finals.

Interestingly enough, I was repeating some voltage readings yesterday and when I put my VOM lead into Pin 1 of Q90, the bottom of my hand - which was touching the IRF510 heat sinks- started to get warm.  A glance toward the wattmeter told me there was RF out!

The TX output quit when I removed the lead and returned more or less when touching the Q90 pin again. After touching the side points for 2cm around in all directions with nothing to show for it, I went and threw in, a 2N2222 as an upgrade, taking care to bend the gate backward betwixt the other two leads so that the E., B., and C. were on the circuit board pads they need to be on.  In firing the rig on, the '2222 gets nice and toasty, as does the pre driver pair (as they did before).

So, did that fix the problem? Actually, no. Still no RF out, now.  I put a 2N3904 back in and still nothing, though, being as the pre drivers are warm as well, I'd suspect the problem comes after them. An RF "sniffer" loop shows that RF fields diminish quickly, downstream of Q90 toward the rest of the transmitter power stages.

I spent the time believing in a Q90 stage solution because of the sporadic full-strength output while taking measurements. It defies explanation how simply making metallic contact can restore seemingly full output, in a way that a re-solder and/or Q90 replacement didn't fix.

I'm sure that reading this report makes no sense. Well, neither did it make sense being there at the time so we're in the same place on that count.

Diving back in after the day job is done with.



-Ted
K3RTA


On Sat, Feb 16, 2019 at 6:00, Raj vu2zap
<rajendrakumargg@...> wrote:
Ted,

does the bias control pots change the idle current ? if it does then finals are fine and final circuit is fine!

Raj

At 16/02/2019, you wrote:
>On Sun, Feb 10, 2019 at 11:11 PM, Raj vu2zap wrote:
>
>
>Your finals are blown. Check if you have TX 5V at the output of regulator, if yes
>then replace IRF510's.
>--------------------------
>
>Nope, new finals made no difference, surprisingly enough.  Have put this aside for a few days to get a fresh start next time diving into this. It's interesting how such a seemingly simple thing gets complicated. 
>
>See my update above for the latest readings, if you like.  I'd love to see this back on line before my V5 package shows up next week :)
>
>
>Regards,
>
>Ted
>K3RTA

Re: load ver.4 original firmware into ubitx ver.3

MVS Sarma
 

While on process of loading a newer revision firmware, it is better to clear the eeprom saved contents.
There is some .ino called  eeprom_clear (Thanks to om mahesh  vu2iia who acquainted me of this). it helps clean up the Arduino nano
Hope it helps.
sarma
vu3zmv
 


On Mon, Feb 18, 2019 at 9:00 PM IZ4SJP Carlo <carlo@...> wrote:
maybe someone will be so kind to describe step by step how to load version 4 firmware into my ubitx version 3?
thanks in advance
IZ4SJP Carlo

Re: load ver.4 original firmware into ubitx ver.3

Jack, W8TEE
 

Sarma:

In some cases, subsequent software programs may read the EEPROM memory space to decide whether it has been written to before and, if so, acts differently based on what it read from EEPROM address 0x00. Most Arduino microcontrollers are "factory set" with an EEPROM value of 0xFF. You should check to see if your program responds differently based on EEPROM values. If so, you may want the .ino program to write that specific value to the EEPROM.

Jack, W8TEE

On Monday, February 18, 2019, 11:15:44 PM EST, MVS Sarma <mvssarma@...> wrote:


While on process of loading a newer revision firmware, it is better to clear the eeprom saved contents.
There is some .ino called  eeprom_clear (Thanks to om mahesh  vu2iia who acquainted me of this). it helps clean up the Arduino nano
Hope it helps.
sarma
vu3zmv
 

On Mon, Feb 18, 2019 at 9:00 PM IZ4SJP Carlo <carlo@...> wrote:
maybe someone will be so kind to describe step by step how to load version 4 firmware into my ubitx version 3?
thanks in advance
IZ4SJP Carlo

Re: load ver.4 original firmware into ubitx ver.3

MVS Sarma
 

Got it  Jack,
But when after loading  , say kd8cec firmware , i had issues and my friend guided me and after that It had worked as expected.

regards
sarma
vu3zmv
 
 


On Tue, Feb 19, 2019 at 10:05 AM Jack Purdum via Groups.Io <jjpurdum=yahoo.com@groups.io> wrote:
Sarma:

In some cases, subsequent software programs may read the EEPROM memory space to decide whether it has been written to before and, if so, acts differently based on what it read from EEPROM address 0x00. Most Arduino microcontrollers are "factory set" with an EEPROM value of 0xFF. You should check to see if your program responds differently based on EEPROM values. If so, you may want the .ino program to write that specific value to the EEPROM.

Jack, W8TEE

On Monday, February 18, 2019, 11:15:44 PM EST, MVS Sarma <mvssarma@...> wrote:


While on process of loading a newer revision firmware, it is better to clear the eeprom saved contents.
There is some .ino called  eeprom_clear (Thanks to om mahesh  vu2iia who acquainted me of this). it helps clean up the Arduino nano
Hope it helps.
sarma
vu3zmv
 

On Mon, Feb 18, 2019 at 9:00 PM IZ4SJP Carlo <carlo@...> wrote:
maybe someone will be so kind to describe step by step how to load version 4 firmware into my ubitx version 3?
thanks in advance
IZ4SJP Carlo

Re: SMD inductors for L5 and L7 replacement

Nigel G4ZAL
 

Hi Peter,

The forum does a good job of obscuring your email address.

I got the prefix part before the @ - what is the suffix part?

I don't subscribe by email as I used to get a gazillion emails in the early days and now only check the web pages I'm interested in via a browser.

Nigel - G4ZAL

----------------------------------------------
Hello Nigel,

peter.russell@...

Let me have your address so I can send 'em.

Peter.

Re: SMD inductors for L5 and L7 replacement

Karl Heinz Kremer, K5KHK
 

You don’t need somebody’s email address to send them a private message via the web interface: Just click on the reply button, but then use the “Private” button below the message editor - it’s gray and on the right side. When you do that, groups.io will send out a normal email, that the recipient can reply to. 
--
Karl Heinz - K5KHK

Re: No TX #ubitx

Ted
 

Jim, 

Unknown at this time, but I'll look when hi.e again. Havi g replaced Q90 with another SMD type 3904 and then a can-sized 2N2222a, the condition is notrepeatable.

The volt readings on the upgraded Q90 are: base 3.1v, emitter 3.4v, and colector 11.8v approx. from memory and the [b] & [e] could be reversed.  Staying up way too late looking at this stuff.


Raj,

I'll take a look at L1 through L4.  Thanks much.



Regards,

Ted
K3RTA

V5 and JackAl

Jack, W8TEE
 

All:

We just got our V5 boards and it appears that there will be no hardware changes needed to the JackAl board. We are working on the software changes, which we don't think will be too extensive. We did look at the 2nd IF filter response and it looks good:
 


We will keep you posted.

Jack, W8TEE

Power instead of S-Meter for CEC-software? #ubitx #smeter #arduino

anders@...
 

Hi all,
I guess it would be possible to measure power output instead of signal strength in the CEC-software, and display in the S-meter. I have copied the Elecraft K2:s RF-detector which gives me between 1 and 4V (~1-10W) that I would like to feed to arduino pin A7. What changes has to be done in the firmware to make this work? Anyone?

Re: Power instead of S-Meter for CEC-software? #ubitx #smeter #arduino

Evan Hand
 

The answer will depend on what other options you have implemented or plan for the CEC run device.  If you are not using or plan to use the stand alone signal analyzer with a second nano, then it should be just a mater of configuring the CEC software as if it were the analog S meter interface.  I have not done this, so would assume that you can adjust the calibration of the display via the configuration or memory manager software from KD8CEC. 

If you plan to use the stand along analyzer, then it would be a little more involved.  I do not have an answer for that at this time.

Please verify the above before making plans or changes, as I am going from memory and have not looked at this for a long period (for me) of time.

73
Evan
AC9TU

Re: SMD inductors for L5 and L7 replacement

Nigel G4ZAL
 

Thanks Karl.

@Peter
Paypal sent and I sent to your email address and it go returned with ... Server error: '452 ... temporary failure'

Nigel

Re: Power instead of S-Meter for CEC-software? #ubitx #smeter #arduino

Jack, W8TEE
 

Check out the map() function. It allows you to map the 1-4V read on the pin to a range of values. The voltage readings are returned as values between 0-1023. You can then map those values to whatever you need.

Jack, W8TEE

On Tuesday, February 19, 2019, 11:47:11 AM EST, anders@... <anders@...> wrote:


Hi all,
I guess it would be possible to measure power output instead of signal strength in the CEC-software, and display in the S-meter. I have copied the Elecraft K2:s RF-detector which gives me between 1 and 4V (~1-10W) that I would like to feed to arduino pin A7. What changes has to be done in the firmware to make this work? Anyone?

Re: SMD inductors for L5 and L7 replacement

Curt
 

I did notice from the v5 schematic that it still uses toroids, not shielded inductors. I am debating with myself a bit how necessary they are. In my own unit, the toroids seem to he okay, only displaying sensitivity on 40 meters, maybe mostly on CW. This has me thinking that my v4 experiences RF coupling into L5 on 7 MHz, where it then gets amplified by the BiDi amp and cancels with the excitation piped into the biased mixer (that then isn't a mixer but a resistive coupler). But why only on 7 MHz? I would guess it may be coupled from PA output. I should try same measurement of power into a dummy load for ssb, evaluating touch sensitivity of L5. My cw output dips a couple watts on 40m only, so its mostly a curiosity.  A visit to spectrum analyzer has been delayed, so no data yet to see if only adding input relays on a daughter board plus a 45 MHz LC filter is enough to cure spurious.

73 Curt