Re: V4 TX problems and new finals

Jerry Gaffke

Don,

So you have temp sensors that turn off when they get hot, and a bunch of extra IRF510's?
An excellent opportunity to better understand what an IRF510 does!

Here's the data sheet:
https://www.vishay.com/docs/91015/sihf510.pdf

Assume we tie the SOURCE pin of the IRF510 to ground, the DRAIN pin to one side of your fan, and the 12v line to the other side of the fan.
When the IRF510 conducts from DRAIN to SOURCE, the fan will be working.

How do we tell the IRF510 to turn on?
We raise the GATE pin from zero volts (measured with respect to the SOURCE pin, which in our case is the same as ground)
up to maybe 12 volts.  You want 5 volts to turn it on good, anything over 20 volts on the GATE can destroy the IRF510.
So if we tie a 1k resistor between the 12v power supply and the GATE, the IRF510 will turn on and the fan will spin up.

How do we get the IRF510 to turn off?
We can now just short that GATE pin to ground with a wire, that puts 12 volts across the 1k resistor and we waste a measly 12v/1k = 12 milliamps of current.

How can we use your normally closed thermal switches to short the gate to ground when the thermal sensor cools off?
Just tie the thermal sensor between the IRF510 GATE and ground.
When the the thermal sensor gets hot, the switch inside opens up, the GATE is no longer shorted to ground
and so the wimpy 1k resistor can pull the gate up to 12v, which turns on the IRF510, which allows current to flow through the fan.

As a bonus, if you ever blow the IRF510's in the uBitx main board, you know where you can find a couple spares.

30 Watts out?
That's way too much.
The rig is probably distorting badly, and you are seriously in danger of destroying your uBitx.

Jerry

On Wed, Dec 18, 2019 at 09:14 PM, Don - KM4UDX wrote:
Then I moved up to +15V on the brown wire. I got more power. Then +20V.  And I got even more power. I was mad with power! How much power you ask? 30W on 80m. And the spectral purity is only sort of marginal. hahaha.  I keep it below 30w...mostly...I try to keep it down...really I do...but sometimes...I ...just...can't...help...myself....ahhhhh.

I felt like Dr. Frankenstein..."it lives!!!"

The hormonally excessive large heat sinks got thermal switches to drive fans. The case got cut to hold the fans. Power was supplied to the fans. I was ready to conquer the world!

Except the fans didn't really work.  The thermal switches were "on" all the time, the fans even with soft mounts made a racket, and the stupid fans would clog every time I hit TX.  And then they died. Maybe cause I paid \$1.99 per fan?

So I need to redo all the fan business.   And I didn't think to test the thermal switches ahead of time. Normally open? Normally closed? hun? what? Such an idiot.

Re: Audio Buzzing

N7XG

I am also having the same buzzing as described. I am using the uBitx with a 5" nexion screen, the base package including the audio amp. I have tried 2 different power supplies for the touch screen, same hum on both.

The hum is constant, that is always at the same level regardless of the volume control.

Re: V4 TX problems and new finals

ajparent1/kb1gmx

Using variable speed fans by running them at 5V for 12V fans and using
temperature sensing to speed them up  works well.   Save for you cannot
touch the heatsinks as they are RF hot so they must be grounded and
use mica or ceramic (alox) pads to isolate the  MOSFETS.

You want to keep the fans on during RX to get the heatsinks cool again
or the next sill start warmer and get hotter still. If done on PTT a timer
on the fans may work.

With any approach a air tight case is going to result in problems.
You need to get the heat outside either by air flow in the case or
by using the case (aluminum) as part of the heatsink with finned
surface outside.
---------------------------------
No direct email, it goes to bit bucket due spam

Re: V4 TX problems and new finals

ajparent1/kb1gmx

Mark,

Running  at 175C is way too hot for any reliability.
Keep it cooler.  That means more heatsink as you pay a
2.5degreeC/per watt of heat dissipated price for the part.
At over 100C you need to seriously derate the part of it
so it will survive.

Reminder at 10W out you doing about 10W of it as heat.
Its not a lot but you have to get it out of the part and that
is harder.  SSB its not that bad for acerage power but at
FT8 full bore, fans!

So to stay cool you need a heatsink that stays much cooler
say under 45C (about 112F) measured as then you have

---------------------------------
No direct email, it goes to bit bucket due spam

Re: V4 TX problems and new finals

Yep I'm sure,

If you read his article about the amp, you can see OM Kosser is a really careful designer that worked very hard to make that amp a nice reliable design.  So he was being prudent and recommending to keep the transistor within recommended temp range of the datasheet.  He did mention that it might be safe running to a maximum junction temperature of 175C which means a heatsink temperature max of 57C.  So that seems to be what some here may be doing.

I was mostly posting the info for users wanting to do digital modes and giving them an idea of what to watch out for.  In that regards, the choice of a turn-on temperature of 35C is not a bad choice at all.  The same thermal switches could be kept and just use a relay or power switching transistor to reverse the mode.

Personally, I think just operating the fans on PTT is a good way to go as long as a larger diameter/lower speed fan is chosen to keep it quiet while keeping the air flow up.

73,

Mark.

Re: V6 radino on V5 board?

Jim Sheldon

Yes. V6 doesn't use an LCD at all. It uses one of the 2.8" TFT color screens and the pinouts go to a 14 pin connector that's mounted down the right side of the V6 Raduino board. V6 Raduino is the same size and has the same mounting holes as the 2.8" ILI9341 display they (and we) use.

If his schematic for the V6 is right, the pinouts for the Display (they are all available on the old 16 pin LCD connector)

I can't get a NANO (with Farhan's factory code) to boot to the 2.8" display (same one they used). Your code doesn't boot to the display either either in a Teensy 4 or NANO. I tried it on both a Raduino clone of ours and a real Raduino (the one from the V5 board you gave me). Neither work.

Jim

Signals from the display and where they are supposed to go according to the Raduino schematic:
1 -- VCC +5V to +5 on the Raduino
2 -- GND to GND on the Raduino
3 -- CS to D10 on the NANO or Teensy
4 -- RESET tied to +5
5 -- D_C to D9 on the NANO or Teensy
6 -- SDI/MOSI to D11 on the NANO/Teensy
7 -- SCK to D13 on the NANO/Teensy
8 -- LED to +5 for full brightness or through 22 ohm resistor for less brightnexx
9 -- SDO/MISO to D12 on the NANO/Teensy
10 -- T-CLK (parallel with SCK) to D13
11 -- T_CS to D8 on NANO/Teensy
12 -- T_DIN parallel with MOSI
13 -- T_DO parallel with MISO
14 -- T_IRQ not used leave open

Re: Templates for China EF01 Electronic Plastic Shell Project Case for a BITX40

KE2GKB

Apologies for the late reply Steve,
Tinkercad lets you export flat drawings for engraving and such. That may suite your needs. However thingiverse i believe has no support for such im afraid.

On Thu, Dec 19, 2019 at 2:17 PM Steve Sawicki via Groups.Io <ab2et=yahoo.com@groups.io> wrote:
Erwin,
I do agree with you that plastic can be a bit brittle that is why I was looking for a template as I did not want to find a insert replacement when a failure did occur.
, I will be very careful as I will be using my Dremel tool with a plastic routing bit or the nibbling tool, no matter which a flat file will be still be needed.
Unfortunately in my new  retirement home shop space and funds for 3D printers are at a premium, thus the old school construction techniques.
I downloaded the panel software and files an I now have a printed overlay that I can use to mark up the panel.
Best Regards.
Best wishes for the Holidays and a Healthy and Happy New Year.
Steve
AB2ET

--
Tim Keller - KE2GKB
https://shop.kit-projects.com

Re: V4 TX problems and new finals

Don - KM4UDX

After lots of empirical observation, the IRF510 heatsink (w/out mica insulator) can get a lot hotter than 32deg Cel.  And live.

Pointing my Lasergrip 774 standoff heat sensor at the TO-220, and at the nearby surrounding heat sink, lets just say we get north of 60°C. on the long 2 minute  fully modulated WSPR mode transmission periods.

Which is good news for the imprudent among us. And it is really cold here in Northern Virginia, so the extra heat is not all bad...

Don
km4udx

Re: Templates for China EF01 Electronic Plastic Shell Project Case for a BITX40

Steve Sawicki

Erwin,
I do agree with you that plastic can be a bit brittle that is why I was looking for a template as I did not want to find a insert replacement when a failure did occur.
, I will be very careful as I will be using my Dremel tool with a plastic routing bit or the nibbling tool, no matter which a flat file will be still be needed.
Unfortunately in my new  retirement home shop space and funds for 3D printers are at a premium, thus the old school construction techniques.
I downloaded the panel software and files an I now have a printed overlay that I can use to mark up the panel.
Best Regards.
Best wishes for the Holidays and a Healthy and Happy New Year.
Steve
AB2ET

Re: V6 radino on V5 board?

W2CTX

I am software but the wiring on the raduino for the LCD
connector is totally different between V5 and V6?

On December 19, 2019 at 11:46 AM "Jerry Gaffke via Groups.Io" <jgaffke=yahoo.com@groups.io> wrote:

From Farhan's post of https://groups.io/g/BITX20/message/73708
The V6 is just an easier to assemble v5. No difference at all but for the touch display.
So the answer is yes, you can update a v5 to have a v6 style display.
No need to swap out the entire Raduino, just the display itself.
Then burn the v6 firmware into the Nano.

I believe the Raduino board has not changed since first introduced for the Bitx40,
except that now a socket is used on the Nano to make it easier to replace.

The old 2x16 display just plugs in, hopefully the new display can plug into
that same old connector with no soldering necessary.

In the past, it has been possible to buy a Raduino separately from hfsignals for about \$30.
I don't see that as an option on hfsignals.com now, the only items to buy are the complete v6 uBitx and the Antuino.
But if the display is easy to swap out, and the Nano is socketed and thus easy to replace if it fails,
that covers most of why you would want a new Raduino.
About the only other stuff on the Raduino is connectors, LM7805 regulator, and the si5351.
If you really do need a new Raduino there have been third parties posting to the forum
offering their own version for sale, some of them with significant enhancements.

Jerry, KE7ER

On Thu, Dec 19, 2019 at 08:18 AM, A Kiddoo wrote:

Will it be possible to purchase a V6 Radino only and adapt it to a V5 main
board? Could this be a possible future upgrade path? Inquiring minds want
to know.

Re: V6 radino on V5 board?

W2CTX

I am software but the wiring of the LCD connector is totally
different between V5 and V6.

On December 19, 2019 at 11:46 AM "Jerry Gaffke via Groups.Io" <jgaffke=yahoo.com@groups.io> wrote:

From Farhan's post of https://groups.io/g/BITX20/message/73708
The V6 is just an easier to assemble v5. No difference at all but for the touch display.
So the answer is yes, you can update a v5 to have a v6 style display.
No need to swap out the entire Raduino, just the display itself.
Then burn the v6 firmware into the Nano.

I believe the Raduino board has not changed since first introduced for the Bitx40,
except that now a socket is used on the Nano to make it easier to replace.

The old 2x16 display just plugs in, hopefully the new display can plug into
that same old connector with no soldering necessary.

In the past, it has been possible to buy a Raduino separately from hfsignals for about \$30.
I don't see that as an option on hfsignals.com now, the only items to buy are the complete v6 uBitx and the Antuino.
But if the display is easy to swap out, and the Nano is socketed and thus easy to replace if it fails,
that covers most of why you would want a new Raduino.
About the only other stuff on the Raduino is connectors, LM7805 regulator, and the si5351.
If you really do need a new Raduino there have been third parties posting to the forum
offering their own version for sale, some of them with significant enhancements.

Jerry, KE7ER

On Thu, Dec 19, 2019 at 08:18 AM, A Kiddoo wrote:

Will it be possible to purchase a V6 Radino only and adapt it to a V5 main
board? Could this be a possible future upgrade path? Inquiring minds want
to know.

Re: ubitx v6, refactored code

Ashhar Farhan

Rick and gang,
See if this helps?

On Thu 19 Dec, 2019, 10:31 PM Jerry Gaffke via Groups.Io, <jgaffke=yahoo.com@groups.io> wrote:
The code of post https://groups.io/g/BITX20/message/44278
makes calibration trivial and also provides a way to center the signal
in the 45mhz and the 12/11mhz crystal filters.

With regard to the v6, see post   https://groups.io/g/BITX20/message/73653
And if curious, follow all the links from that post down into previous discussions.

Jerry, KE7ER

On Wed, 18 Dec 2019 at 21:58, Ashhar Farhan <farhanbox@...> wrote:

There is no problem with the firmware on the shipped units. They wouldn't work at all if the firmware didnt compile.
A bracket was missing on the uploaded firmware as the person cleaned up a commented line under the if statement before uploading the source code. You don't have to load or tune up the radios and they work out of the box.
I understand that rick had a problem with a badly crimped connector. It would be impossible for us to test all the individual components. We test the displays, arduinos, all coils and crystals before the board is assembled. After assembly, we set the PLL calibration, the bfo, the pa bias, we test ssb and cw on 80, 40, 20 and 10 meters. The alignment has to be done on the pair of the digital board (raduino) and the analog board (the main board) as the crystals of the IF filter and the crystal of the Si5351 have to be calibrated and aligned with respect to each other too.
I've stumbled across a couple posts from kd8cec about his experiments
with IF offset where he found the si5351 calibration would also
benefit offsetting the first IF (45MHz) as well to compensate for both
what I guess may be a little drift between 45MHz filters but also the
si5351 xtal.

I'm experimenting with it on my v4 ubitx and see that indeed there's a
benefit from offsetting the first IF / VFO frequency too. I'm trying
to nail it down using the kd8cec firmware and then port the minimum
set of changes needed to your v4 firmware.

Have you looked into that as well?

Thanks!

Re: For Sale new uBITX kit with additional uBITX board plus display

Jonas Sanamon

Hi Patrick!

We can split the parts between us,  If You take the extra board for \$50, and I take the new unbuildt complete kit for \$100, we'd have a win-win.
Seller ships all of it to You and You then ship to me?  I'll pay You for the shipping US-Sweden, and can send You the \$100 up-front.

Agree?

Best regards
Jonas - SM4VEY

Den tors 19 dec. 2019 kl 16:05 skrev Patrick Pelletier <ve9pax@...>:

Hi Jim,

Would you be willing to sell the board only?  I have a Raduino with a board I messed up...

Thanks,

Patrick
VE9PAX

On Thu, Dec 19, 2019, 10:59 AM Jim Sorenson, <kjsorenson@...> wrote:
I have a new uBITX kit and a spare uBITX board for sale.

The uBITX board without the Raduino was part of a working system till I blew out the Raduino.
The uBITX kit is new in the box and was purchased within the last 5 months.
I'll include a display and various part from the first kit.

\$150 for both kit and additional board, parts and display
PayPal to my email and I can ship today CONUS - shipping on me.

I enjoyed using the uBITX while the Raduino was OK, but am moving house and would not get back to this project for some time.

Many thanks,

Jim Sorenson
W3BH

www.w3bh.com

Re: V4 TX problems and new finals

Hi,

In WA2EBY's original paper he carefully studied the requirements for keeping the IRF510 at a safe temperature:

"The amplifier can also be damaged by overheating.  This limitation is imposed by the TO-220 packages in which Q1 and Q2 are housed.   The thermal resistance from junction to case is a whopping 3.5°C/W.  This huge value makes it virtually impossible to keep the junction temperature from exceeding the +150°C target for good reliability."

"
To keep the transistor junction temperature below +150°C requires preventing the transistor case temperature from exceeding 46.8°C."

"Also, there is a temperature rise across the mica insulator between the transistor case and heat sink of 0.5°C/W.  That makes the maximum allowable heat sink temperature limited to 46.8 – (0.5 × 29.5) =32°C."

73,

Mark

Re: V4 TX problems and new finals

Don - KM4UDX

Edwin -- why did get those exact thermal switches? Because I wanted the fans on, until it got warm, then turn off. And that worked perfectly. hahah.

No, actually, I was a dope and didn't pay enough attention when I purchased them, and then I failed to test'em.  [We don't need no stinking test'en here!!]

So the switch and fan combination worked the exact opposite way I wanted. Perfectly wrong.

On the other hand, they quickly failed. So I guess we are even.

At the risk of further humiliation, would you pass me a link to the correct thermal switch?  I will endeavor to not spectacularly screw up the second go-round.

Don
km4udx

Re: Templates for China EF01 Electronic Plastic Shell Project Case for a BITX40

Erwin Serlé

Steve, plastic might be more brittle than metal, so try out the knibbling on a scrap piece first.

I am currently printing on my Ultimaker Original and also building a Dremel 4000 tool 3D printed CNC machine (Nikodem Bartnik), to make pcb's and panels and so on
--
PE3ES - F4VTQ - Erwin
73

Re: V4 TX problems and new finals

Erwin Serlé

On Wed, Dec 18, 2019 at 09:14 PM, Don - KM4UDX wrote:
Don
km4udx
I see you use NO switches starting from 35 degrees Celsius. In a case that would mean the fans are almost always on. Why did you pick this specific temperature point? I just bought a 50 degrees Celsius to switch a fan in a power supply on. For uBITx it might be somewhere in the 50-60 degrees range I would think.

PE3ES/F4VTQ-Erwin

--
PE3ES - F4VTQ - Erwin
73

Re: ubitx v6, refactored code

Jerry Gaffke

The code of post https://groups.io/g/BITX20/message/44278
makes calibration trivial and also provides a way to center the signal
in the 45mhz and the 12/11mhz crystal filters.

With regard to the v6, see post   https://groups.io/g/BITX20/message/73653
And if curious, follow all the links from that post down into previous discussions.

Jerry, KE7ER

On Wed, 18 Dec 2019 at 21:58, Ashhar Farhan <farhanbox@...> wrote:

There is no problem with the firmware on the shipped units. They wouldn't work at all if the firmware didnt compile.
A bracket was missing on the uploaded firmware as the person cleaned up a commented line under the if statement before uploading the source code. You don't have to load or tune up the radios and they work out of the box.
I understand that rick had a problem with a badly crimped connector. It would be impossible for us to test all the individual components. We test the displays, arduinos, all coils and crystals before the board is assembled. After assembly, we set the PLL calibration, the bfo, the pa bias, we test ssb and cw on 80, 40, 20 and 10 meters. The alignment has to be done on the pair of the digital board (raduino) and the analog board (the main board) as the crystals of the IF filter and the crystal of the Si5351 have to be calibrated and aligned with respect to each other too.
I've stumbled across a couple posts from kd8cec about his experiments
with IF offset where he found the si5351 calibration would also
benefit offsetting the first IF (45MHz) as well to compensate for both
what I guess may be a little drift between 45MHz filters but also the
si5351 xtal.

I'm experimenting with it on my v4 ubitx and see that indeed there's a
benefit from offsetting the first IF / VFO frequency too. I'm trying
to nail it down using the kd8cec firmware and then port the minimum
set of changes needed to your v4 firmware.

Have you looked into that as well?

Thanks!

Re: V6 radino on V5 board?

Jerry Gaffke

From Farhan's post of    https://groups.io/g/BITX20/message/73708
>  The V6 is just an easier to assemble v5. No difference at all but for the touch display.
>  You can buy the tft off ali express for about 5 dollars and upgrade the firmware.

So the answer is yes, you can update a v5 to have a v6 style display.
No need to swap out the entire Raduino, just the display itself.
Then burn the v6 firmware into the Nano.

I believe the Raduino board has not changed since first introduced for the Bitx40,
except that now a socket is used on the Nano to make it easier to replace.

The old 2x16 display just plugs in, hopefully the new display can plug into
that same old connector with no soldering necessary.

In the past, it has been possible to buy a Raduino separately from hfsignals for about \$30.
I don't see that as an option on hfsignals.com now, the only items to buy are the complete v6 uBitx and the Antuino.
But if the display is easy to swap out, and the Nano is socketed and thus easy to replace if it fails,
that covers most of why you would want a new Raduino.
About the only other stuff on the Raduino is connectors, LM7805 regulator, and the si5351.
If you really do need a new Raduino there have been third parties posting to the forum
offering their own version for sale, some of them with significant enhancements.

Jerry, KE7ER

On Thu, Dec 19, 2019 at 08:18 AM, A Kiddoo wrote:
Will it be possible to purchase a V6 Radino only and adapt it to a V5 main board? Could this be a possible future upgrade path? Inquiring minds want to know.

Re: ubitx v6, refactored code

On Wed, 18 Dec 2019 at 21:58, Ashhar Farhan <farhanbox@...> wrote:

There is no problem with the firmware on the shipped units. They wouldn't work at all if the firmware didnt compile.
A bracket was missing on the uploaded firmware as the person cleaned up a commented line under the if statement before uploading the source code. You don't have to load or tune up the radios and they work out of the box.
I understand that rick had a problem with a badly crimped connector. It would be impossible for us to test all the individual components. We test the displays, arduinos, all coils and crystals before the board is assembled. After assembly, we set the PLL calibration, the bfo, the pa bias, we test ssb and cw on 80, 40, 20 and 10 meters. The alignment has to be done on the pair of the digital board (raduino) and the analog board (the main board) as the crystals of the IF filter and the crystal of the Si5351 have to be calibrated and aligned with respect to each other too.
I've stumbled across a couple posts from kd8cec about his experiments
with IF offset where he found the si5351 calibration would also
benefit offsetting the first IF (45MHz) as well to compensate for both
what I guess may be a little drift between 45MHz filters but also the
si5351 xtal.

I'm experimenting with it on my v4 ubitx and see that indeed there's a
benefit from offsetting the first IF / VFO frequency too. I'm trying
to nail it down using the kd8cec firmware and then port the minimum
set of changes needed to your v4 firmware.

Have you looked into that as well?

Thanks!