Topics

Homebrewing a uBITX according to M0HYT


Marko Barovic
 

Hi all! Thanks for having me in this great group of people!

I'm a ham for some years now but I don't have much experience in any aspect. Since I'm a medical doctor and not really formally educated in electronics and like to tackle a big challenge to make me learn stuff, I decided to build a uBITX from scratch according to the plans of M0HYT and based on his PCB design. Now, my PCB is getting made and I'm starting to source the parts I will need.

I came accross the trifilar wound toroids that need to be wound by hand apparently. That is of course no problem, I am excited to learn about absolutely any part of the project I will need to. However, I'm a bit confused by the schematic as to what I need to do and I can't really find the answer on the internet.

 

So this is the toroid I'm supposed to wind. Does this mean that I wind three turns trifilar and then continue up to 8 turns with only one wire?
Also, if there's a universally accepted toroid core people are using, I would be very thankful to know about it!

Thanks a lot in advance!

73


Evan Hand
 

The same basic circuit for the HFSignals uBiTx has 10 trifilar turns on the toroid.  Here is a picture of a number of them on the board:

Hope this helps.
73
Evan
AC9TU


Joe Puma
 

Wow what a undertaking for someone who is not even an EE and you seem like you already are familiar.  Bravo 👏🏼 👏🏼👏🏼👏🏼

Joe
KD2NFC 



On Mar 25, 2020, at 2:18 PM, marko.barovic.me@... wrote:

Hi all! Thanks for having me in this great group of people!

I'm a ham for some years now but I don't have much experience in any aspect. Since I'm a medical doctor and not really formally educated in electronics and like to tackle a big challenge to make me learn stuff, I decided to build a uBITX from scratch according to the plans of M0HYT and based on his PCB design. Now, my PCB is getting made and I'm starting to source the parts I will need.

I came accross the trifilar wound toroids that need to be wound by hand apparently. That is of course no problem, I am excited to learn about absolutely any part of the project I will need to. However, I'm a bit confused by the schematic as to what I need to do and I can't really find the answer on the internet.

 
<Annotation 2020-03-25 173356.png>


So this is the toroid I'm supposed to wind. Does this mean that I wind three turns trifilar and then continue up to 8 turns with only one wire?
Also, if there's a universally accepted toroid core people are using, I would be very thankful to know about it!

Thanks a lot in advance!

73


Marko Barovic
 

Thanks for the input, Evan! I saw them on the schematic of the HFSignals device. You think I'll be fine winding 10 trifilar turns on the cicruit?

73
Marko 4O6MAG


Marko Barovic
 

Thanks for the words of encouragement, Joe! I'm counting on communities like this to push me when I get stuck :)
I'm going on the line that for whatever circuit I don't understand, I will just have to learn how it works and more knowledge is never bad! I think this is a good thing for me to learn the inner workings of a radio so that when I start earning actual money and get something more "out of the box", I am not only blindly pressing the knobs but actually understand what is going on.

73 de 4OMAG


 

Marko,
There is an outstanding video on You Tube by W2AEW Alan Volke on winding trifilar toroids
https://youtu.be/a8ViWS61hsU

In general, all the videos by him are highly recommended.

73 de k1jbd
bammi


Joe Puma
 

Your determination and this group will get you through it all 😊. Looking forward to seeing the results.

 

 

73

Joe

KD2NFC

 

Sent from Mail for Windows 10

 

From: marko.barovic.me@...
Sent: Wednesday, March 25, 2020 3:05 PM
To: BITX20@groups.io
Subject: Re: [BITX20] Homebrewing a uBITX according to M0HYT

 

Thanks for the words of encouragement, Joe! I'm counting on communities like this to push me when I get stuck :)
I'm going on the line that for whatever circuit I don't understand, I will just have to learn how it works and more knowledge is never bad! I think this is a good thing for me to learn the inner workings of a radio so that when I start earning actual money and get something more "out of the box", I am not only blindly pressing the knobs but actually understand what is going on.

73 de 4OMAG

 


Rick Price
 

From the HFSignals Website description
 

Coil Details

L5,  L7 : 12 turns on T30-6
  • L1, L2, L3, L4, L11, L12, L13 : 9 turns on T30-6
  • L14, L15, L16:  10 turns on T30-6
  • L14, L15, L16:  14 turns on T30-6
  • L20, L21, L22: 19 turns on T30-6
All the RF transfomers are 8-10 trifilar turns on FT37-43
 
Rick
KN4AIE


From: BITX20@groups.io [mailto:BITX20@groups.io] On Behalf Of marko.barovic.me@...
Sent: Wednesday, March 25, 2020 3:06 PM
To: BITX20@groups.io
Subject: Re: [BITX20] Homebrewing a uBITX according to M0HYT

Thanks for the words of encouragement, Joe! I'm counting on communities like this to push me when I get stuck :)
I'm going on the line that for whatever circuit I don't understand, I will just have to learn how it works and more knowledge is never bad! I think this is a good thing for me to learn the inner workings of a radio so that when I start earning actual money and get something more "out of the box", I am not only blindly pressing the knobs but actually understand what is going on.

73 de 4OMAG


Evan Hand
 

Yes, since it is a transformer, it is the turns ratio that is most important, though the number of turns can impact the frequency response along with the core material.  I do not have the core material, so I assume that you were able to get that somewhere else.  For HF transformers it is normally powdered iron with size according to power. requirements.

Here is a YouTube on winding a trifilar transformer.  Not sure that the core material that he is using is the same as that used by HFSignals for the transformer.  As you can see in the picture, the transformers are black/clear, not yellow/black as are the type 43 shown.  I do not know the specifications of the HFSignals transformers.

Sorry, I can not be more specific.
73
Evan
AC9TU


Marko Barovic
 

Thank you all very much! Cheers and until the next time I get stuck on something!

I am amazed at how friendly and helpful this group of people is!

Cheers and 73
Marko

On Wed, Mar 25, 2020 at 8:47 PM Evan Hand <elhandjr@...> wrote:
Yes, since it is a transformer, it is the turns ratio that is most important, though the number of turns can impact the frequency response along with the core material.  I do not have the core material, so I assume that you were able to get that somewhere else.  For HF transformers it is normally powdered iron with size according to power. requirements.

Here is a YouTube on winding a trifilar transformer.  Not sure that the core material that he is using is the same as that used by HFSignals for the transformer.  As you can see in the picture, the transformers are black/clear, not yellow/black as are the type 43 shown.  I do not know the specifications of the HFSignals transformers.

Sorry, I can not be more specific.
73
Evan
AC9TU



--
Marko Barovic, MD
PhD student, Solimena lab
DZD-Paul Langerhans Institute Dresden of the Helmholtz Zentrum München at TU Dresden


 

A bit of mix up there,

HF transformers, unless they are for tuned designs, are normally of ferrite material.  They are normally clear-coated, and most often type 43 material unless for the upper end of the HF spectrum and then 61 is used more often.

Powdered iron cores are normally for tuned and not wideband use:  their Al value is more precise and does not change a lot with temperature.  Type 6 (yellow coated) material or type 2 (red coated) are most often used in HF.

Note: on some of the bitx transformers, they are wound trifilar but only two windings are used.  So some can be wound bifilar (if you want to save on wire and lessen the chance of getting the ends mixed up).

73,


Mark


 

"They are normally clear-coated..."  Oops, I should have wrote, "They are normally not coated.."


Marko Barovic
 

Hello all! Sorry if resurrecting an old topic is against the customs of this group (I know that on some forums it's frowned upon).

I built up the radio on the PCB, soldered all the components except for the IRF510 (I wanna make sure the receive works and that I didn't make somedumb mistakes with the circuits that would fry my finals before installing them).

Now before connecting it to power, I measured about 1.5kOhm between the - and the + on the power connector. Is this supposed to be like this or should I go looking for a short somewhere?

Thanks in advance and 73!
Marko


Gerard
 

Hello,
I made one as well. It works very well in reception. I haven’t cabled the whole broadcast part, because I only do reception, but it should work. 
cdt


Evan Hand
 

On Mon, Feb 15, 2021 at 03:49 AM, Marko Barovic wrote:
Now before connecting it to power, I measured about 1.5kOhm between the - and the + on the power connector. Is this supposed to be like this or should I go looking for a short somewhere?
Marko,
Gerard might be a better source for information on a homebrew µBITX, as he has built one.  

I did go back and read the HFSignals' "Wire Up" page and found the startup values for the board current is between 90 and 110 ma.  1.5k ohms would be about 80ma, so you should be good to go.

Here is the link to the wire up page:
https://www.hfsignals.com/index.php/ubitx-wire-up/

Good luck and have fun with your build!
73
Evan
AC9TU


Marko Barovic
 

As always, thanks a lot for the fast replies and kindness to both of you! I will snatch some time on the weekend to do an ugly wire up and power it up.

Evan, additionally, thank you for that principle of estimating if the R I measured is ok. As I have literally 0 engineering background that perfectly logical way of checking that never even crossed my mind! I guess I will remember it for good now :)

Thanks a lot again!

73,
Marko

On Mon, Feb 15, 2021 at 11:25 AM Evan Hand <elhandjr@...> wrote:
On Mon, Feb 15, 2021 at 03:49 AM, Marko Barovic wrote:
Now before connecting it to power, I measured about 1.5kOhm between the - and the + on the power connector. Is this supposed to be like this or should I go looking for a short somewhere?
Marko,
Gerard might be a better source for information on a homebrew µBITX, as he has built one.  

I did go back and read the HFSignals' "Wire Up" page and found the startup values for the board current is between 90 and 110 ma.  1.5k ohms would be about 80ma, so you should be good to go.

Here is the link to the wire up page:
https://www.hfsignals.com/index.php/ubitx-wire-up/

Good luck and have fun with your build!
73
Evan
AC9TU



--
Marko Barovic, MD
PhD student, Solimena lab
DZD-Paul Langerhans Institute Dresden of the Helmholtz Zentrum München at TU Dresden


Ashhar Farhan
 

the circuit that you posted in the original post is from the transformer used on the PCB. it is a crazy wire-up because the transformer is mounted on a 6 pin DIP base. This is done so that the women who wind the transformers, can mount it on a base and plug it into an arduino circuit that tests for continuity etc. Then, the assembly team just drops it onto the PCB. 
When you build it, it might make sense to redraw it just like any other diode mixer transformer. there are several source that show you how it is done.
- f

On Mon, Feb 15, 2021 at 4:50 PM Marko Barovic <marko.barovic.me@...> wrote:
As always, thanks a lot for the fast replies and kindness to both of you! I will snatch some time on the weekend to do an ugly wire up and power it up.

Evan, additionally, thank you for that principle of estimating if the R I measured is ok. As I have literally 0 engineering background that perfectly logical way of checking that never even crossed my mind! I guess I will remember it for good now :)

Thanks a lot again!

73,
Marko

On Mon, Feb 15, 2021 at 11:25 AM Evan Hand <elhandjr@...> wrote:
On Mon, Feb 15, 2021 at 03:49 AM, Marko Barovic wrote:
Now before connecting it to power, I measured about 1.5kOhm between the - and the + on the power connector. Is this supposed to be like this or should I go looking for a short somewhere?
Marko,
Gerard might be a better source for information on a homebrew µBITX, as he has built one.  

I did go back and read the HFSignals' "Wire Up" page and found the startup values for the board current is between 90 and 110 ma.  1.5k ohms would be about 80ma, so you should be good to go.

Here is the link to the wire up page:
https://www.hfsignals.com/index.php/ubitx-wire-up/

Good luck and have fun with your build!
73
Evan
AC9TU



--
Marko Barovic, MD
PhD student, Solimena lab
DZD-Paul Langerhans Institute Dresden of the Helmholtz Zentrum München at TU Dresden


Marko Barovic
 

Dear all,

First of all, thank you Ashhar for the input. I am really learning a lot here!

I installed everything, did very ugly wire up and connected the board to a repurposed ATX PSU. I have a bad antenna (the "noise cancelling loop" from swling blog) and the PSU is probably also bad in terms of noise. That said, I do receive noise, on certain frequencies continuous tones that are probably some noise from within my apartment or from somewhere nearby. I know that's not really of much use, but I am so happy and wanted to share here. I am happy no white smoke left the machine when powered on (hi) and I am happy that my radio in broadest terms works and even more happy that I actually understand what's behind it all and that I built it, of course with the immense help of the material shared on this group and the constructors.

Now I need to find a way to calibrate the thing. I have another SI5351A and another nano that I could practice my coding on and make it output something so that I can use that as a reference. Do you think that would be a good idea or it would suffer from the same kind of drift or whatever it's called that the SI5351A that's already in the radio has?

Also, I think I need to find a battery that can power this, I have a feeling the PSU is injecting a ton of noise...

I'm excited to hear any suggestions you all might have.

Thanks a lot again!

73,
Marko


Rafael Pinto [PU1OWL]
 

Marko,

A different Si5351 would not be enough to calibrate. What I did was to get a GPS uBlox Neo M8M module, and set its PPS output to 5MHz. To the PPS pin, I added a short piece of wire and placed over a short piece of wire connected to the uBitx, so to act as a capacitive coupling/attenuator

It worked pretty well.

But what really worked for me was to tune by ear to a strong local AM radio station at 1130kHz, then I moved to our national time signal (1KHz at 10MHz) and adjusted again to get the 1kHz bip right. 

In my uBitx, the calibration was 201275, which corresponds to 230Hz shift on the crystal. BTW, wait a few minutes before starting to tune, as the oscillator drifts a little on the first 10 minutes or so. Go figure...

73, 

Rafael


Curt
 

Marko

I used same method as Rafael, tune in a strong am signal for equal clarity and pitch on LSB  and USB. 

You need a linear power supply, or one designed with extra filtering for radio use. Some ancient wall warts that are rather heavy are likely linear and might at least have enough current for receive.  Check also the usual retail websites that do business in your part of the globe.

Curt