uBITX Transceiver semi-kit build

Discussion in 'Homebrew and Kit Projects' started by KD8UYQ, Aug 14, 2018.

ad: L-HROutlet
ad: l-rl
ad: Left-2
ad: MessiPaoloni-1
ad: Left-3
ad: L-MFJ
ad: l-assoc
ad: Subscribe
  1. KD8UYQ

    KD8UYQ Ham Member QRZ Page

    Hi Art,

    How are sir. The decision to go with only one matching transformer on the output of the 45 Mhz crystal filter was based on the research that Warren Allgyer (WA8TOD) did.

    Here is what he had to say:

    "Here is what I did. I did not try with a second transformer because the results were good enough. After the IMD measurements I would be even more concerned about adding a second transformer because the Q22 emitter follower is operating at an impedance of about 400 ohms and, as such, is one of the major contributors to the IMD issue. Adding a low impedance transformer to this source is likely to make that problem worse. But I have not tested it so I cannot be sure".


    Now the reason for the over all decrease in power output on 40 meters on up is due to the setting of RV1 (the trimmer potentiometer used to adjust the drive level). I found that this pot was turned all the way to the maximum counter clockwise position. In effect the drive level was wide open. I never checked its setting prior to replacing the two pre-driver transistors which is an over sight on my part. With RV1 set to maximum gain the 80 meter band was topped out at 14 watts and the other bands power output significantly higher as reported earlier. I adjusted the drive control pot back until my power output meter just started to move down from the 14 watts indicated then back to the point where the output would go no higher. That turned out to be about 40% of RV1's full travel. With the drive level set properly at 80 meters the output power on the other bands came down but so did the harmonics and spurious signals on 10 meters making the uBITX FCC compliant in every aspect except for that one SSB spurious signal on 10 meters.

    My net results are now in line with what Allison found and published in her "RF and Gain" paper that I provided a link to in my previous post.


    PS: Art, your TS-950 that you dropped off last week is next up on the workbench after I finish up on the TR-7 and IC 746 that I am working on now.
    KG5RKP likes this.
  2. W4ATO

    W4ATO Ham Member QRZ Page

    I saw Warrens , but a few others were making boards for the two transformer models. Good to know i only have to make one.
    Regarding the PA, I read Allisons notes on max bias of 20 ma and thought gain could be an issue. I also asked her about her post of using a single or dual diode to make a AGC, but she said that it would not work on this boards design without major changes. So on to ND6T
    design. Now to create a bom for all these mods.
    No hurry on 950. Once you get it running play with it for awhile. Neat radio. 73
  3. KK5R

    KK5R Ham Member QRZ Page

    This may be old info for this forum but to make it easy to use a speaker for the audio output, I used an AC powered computer speaker. It has a lot of volume so I use the computer speakers at full volume and only use the rig's volume control. Depending on the quality of the computer speakers, you may not like it that much for audio quality but for communications use, they seem fine. I will try some that are not amplified, which most of the AC powered speakers are, to see if the rig has enough power to hear when in the field. I suspect they will also work as well and be better when out in the field. The impedance of the simple outboard speakers may be a factor, though.

    Bob — KK5R
  4. KK5R

    KK5R Ham Member QRZ Page

    Too late to edit/add info to my post above so I'll post a reply.

    Tried the computer speakers without amplification and they work. The audio level out of the speakers is not very high but is adequate, in my opinion. However, because they are simple/cheap speakers, the audio quality was not very good (too tinny) so changed back to the amplified speaker which is one of the two speakers taking drive from both inputs, the second channel is not used since the drive from the radio is adequate.

    Now listening to WTAM 1100 AM and was listening to WLAC 1510 AM last night. It's on LSB so I put it on the exact freq which nulled the carrier. Sounds very good. Cannot use transmitter until I get a good mic since I "probably" burned out the electret mic element by getting it too hot when soldering. The flashlight body for the mic from a previous comment is a good idea and I'll try this. Wondering if a HI-Z magnetic mic would work...
  5. KD8UYQ

    KD8UYQ Ham Member QRZ Page

  6. KD8UYQ

    KD8UYQ Ham Member QRZ Page

    It's been a while,

    The holidays are behind me, this is the time of year where it is pretty quiet at the Radio Museum, and my workbench is clear. A perfect time to take out the uBITX test bed and try the next modification. My uBITX is now FCC compliant except for this one pesky spur up on 10 meters. The 45 MHz crystal filter helped a lot with overall spurious signal suppression but there is one in particular around 12 MHz that bounces above the -43 dB threshold.

    10m spurs.jpg

    A more recent recommended fix for spur suppression was to remove L5 and L7 and replace them with 1210 .68uH surface mount inductors. This experiment was to test and see if this fix in addition to the 45 MHz crystal filter that is already in place would further reduce spurs on 10 meters without adverse side effects. First things first, pull L5 and L7.

    L5 and L7 removed.JPG

    The SMD .68uH inductors that I choose came from Mouser per a link provided by one of the posters on the BITX20 users group. Those were placed on the board per Raj, vu2zap directions and photo.


    Now please note that picture above is of his board not mine (I forgot to take a picture after I installed my Mouser SMD inductors). The Mouser components are a bit smaller and gray not yellow.

    Before I show you the results after L5 and L7 were replaced an explanation as to the test set up itself. Previously I was using my hand mic placed in front of a speaker that was in turn connected to a little two tone generator board that I had on hand to do the spurious signal testing. It worked (kinda) but wasn't very precise.

    This time I used my Rigol dual channel arbitrary waveform generator to give me a calibrated signal injected directly into the microphone input of the uBITX. I set Channel A for 700 Hz and Channel B for 1900 Hz and coupled both channels together so that I could change the amplitude in unison. The output was checked and set on the scope for maximum power output without flat-topping the signal. 28 mv-rms in gives me 5 watts out on 28.850 MHz with a nice sting of Perls looking waveform.

    Using this setup I took a screen shot of the spectrum analyzer before as you have seen above and after; below.

    10m 680uh smd.jpg

    You can see that the spurs dropped in power but so did my output on 28.850 MHz. In fact my output power went from 5 watts to about 200 miliwatts. That's not good.

    To see if this was anomaly due to some adverse interaction between the 45 MHz crystal filter spurious signal mod done early and this one with the L5 & L7 replacement I took my other uBITX main board with no mods done on it at all and swapped out the L5 and L7 on it with the SMD inductors and ran the same before and after tests. The results were the same. Power out goes down to almost nothing so clearly these components that I bought from Mouser do not work for this application. Back to the drawing board I found the exact same SMD devices that Raj used and ordered 10 of them. That's the good news. The bad news is they are coming from the Netherlands so it will be a while before I can try this test again.

    Until next time.

    N3REN likes this.
  7. N1OOQ

    N1OOQ Ham Member QRZ Page

    Seems like the toroidal inductors would be higher Q... But then, I have no idea what they do in the circuit!
  8. KD8UYQ

    KD8UYQ Ham Member QRZ Page

    Hello Thomas.

    I would think so too but apparently swapping out the torroids with surface mounted inductors helps to reduce the spurs generated in the 45 MHz mixer stage of this transceiver. If you scroll up this page to my post #91 you can see the 45 MHz mixer circuit to the right of the schematic depicted there.

  9. AJ8MH

    AJ8MH XML Subscriber QRZ Page

    Looks like you bypassed C211 on the board pic. OOPS, I just read the article on the uBITx page...
    Last edited: Jan 11, 2019
  10. N1OOQ

    N1OOQ Ham Member QRZ Page

    Hmpf... Interesting. I've never seen a single xtal filter in that configuration. Some kind of narrow BW impedance transformation on each side?

Share This Page

ad: AbAuRe-1