Keyboard Shortcuts
Likes
Search
VFO/BFO selecting outputs
HI All,
Getting ready to convert over to using the si5351 for the BFO. I'd like to use rg-174 for both the VFO and BFO lines. The Si5351 has three outputs. The first is already routed on the PCB to the SIP header making it inconvienent to solder coax to it. Is there any reason that I can't use the two other Si5351 outputs that aren't used? That would be easy to solder to. And if so do I change the sketch like: #define VFO_A 0 #define VFO_B 1 to #define VFO_A 1 #define VFO_B 2 I've never done Arduino before, but that looks like it to me. Anything else? Jonathan - KK6RPX |
All three Si5351 outputs are routed from the Si5351 to the Raduino's P3 connector through 0.1uF DC blocking caps. ?You need those DC blocking caps. ?Perhaps you are using a different Si5351 breakout board? ?I believe the Bitx40 currently uses CLK2 for the VFO, leaving CLK0 and CLK1 available. ?I suggest you use CLK0 for the BFO. ?I've seen reports that there is more crosstalk between channels when using CLK1. ? You should be able to find skinnier and more flexible coax out there than RG-174 if that's an issue. ?The impedance (50 or 75 ohms) should not matter much at these frequencies for these frequencies (below 50 mhz) on coax pieces of just a foot or so in length. ?It's possible that the volume control and microphone input could benefit from coax also. On Sat, Apr 8, 2017 at 12:35 pm, Jonathan Peakall wrote:
Getting ready to convert over to using the si5351 for the BFO. I'd like to use rg-174 for both the VFO and BFO lines. The Si5351 has three outputs. The first is already routed on the PCB to the SIP header making it inconvienent to solder coax to it. Is there any reason that I can't use the two other Si5351 outputs that aren't used? That would be easy to solder to. ? |
开云体育No, it's the raduino board. I probably
wasn't very clear. All of the clock outputs do go to the PCB holes
for the SIP header. But only the first 5 pins of the header
actually have pins in them, the rest are open.? I'll have to find
which is which (0,1,2) but one clock output is in the middle of
the 5 pins that are soldered into the PCB. The others go to blank
holes which I could easily solder into with the RG-174. Hope that
makes it clear as mud!
BTW, what coax is smaller and more flexible that RG-174? Got a number? Thanks for the help! Jonathan On 4/8/2017 12:45 PM, Jerry Gaffke via Groups.Io wrote:
|
Ron VE8RT
I'm using Harbour Industries M17/93-RG178 purchased from
www.therfc.com There are likely other sources for RG178 Ron VE8RT On Sat, 08 Apr 2017 13:02:31 -0700 "Jonathan Peakall" <jpeakall@...> wrote: No, it's the raduino board. I probably wasn't very clear. All of the -- Ron VE8RT <ve8rt@...> |
Ron VE8RT
On the header connectors the contacts were extracted and re-used
with the RG178. As I had spare contacts on hand new ones were used if required. The antenna and LO wires were both replaced with RG178. The photo may give you an idea of how small it is. Ron VE8RT On Sat, 08 Apr 2017 13:02:31 -0700 "Jonathan Peakall" <jpeakall@...> wrote: No, it's the raduino board. I probably wasn't very clear. All of the-- Ron VE8RT <ve8rt@...> |
We are using coax to shield the critical signals, not so much as transmission line. ?The coax shield on both ends should be tied to a suitable ground. ?If used on the volume pot (I have no idea if that will help in your particular situation), you probably need two pieces of coax because there are two critical signals. According to this webpage, RG-174 is 2.55mm in diameter, RG-178 is is the smallest listed at 1.8mm: ?? ? The RG series spec was last updated in 1962. ?I have some nice flexible 1.5mm coax on hand, no identifying marks and no idea what it might be called. If you don't want to mess with coax, twisted pairs are a good second best. ?Just twisting the wire provided by hfsigs will help, but better to use smaller diameter wire as that allows the wires to be closer together. ?Twist a ground wire with the critical signal wire, perhaps one or two turns per inch. ?In the case of the volume pot, could just twist all three wires together since we don't care if the two signal wires interact. On Sat, Apr 8, 2017 at 12:45 pm, Jerry Gaffke wrote:
?The impedance (50 or 75 ohms) should not matter much ? |
开云体育Thanks for the coax info. The 174 I
have is small enough for the job, although I'll get some of the
178. However, my real question was about how to change the code to
change which Si5351 output I use.
Also, looking for how to enable the raduino BFO. Thanks! On 4/8/2017 3:32 PM, Jerry Gaffke via Groups.Io wrote:
|
OK, I'll take a stab at this. ?But haven't compiled and tried it, might be sending you off with a bug or two. The VFO_A and VFO_B defines are just an integer used to remember what frequency we are currently operating on. ?This is not fully implemented. ?If it were, you could push a button and flip from talking to your friend at 7.175 with VFO_A and see what the traffic net is doing at 7.270, for example. ?Then push the button again and get back to your friend at 7.175. ?Twiddling the tuning knob changes the frequency of whatever VFO you have currently selected. ?There is of course only one VFO, and it always comes out of channel 2 of the Si5351. ?The Nano sketch is remembering those two VFO frequencies, and toggling the Si5351 frequency out of channel 2 when you press the button. ? In summary, your on the wrong track. Let's assume we leave the VFO at Si5351 channel 2 and put the BFO at Si5351 channel 0. ?Channel 1 is left unused, and is turned off. ?You can probably figure out how to add a piece of coax to channel 0, perhaps just solder it directly into the back of the Raduino board. Add this code to the bottom of the setup() function in the sketch: //si5351.drive_strength(SI5351_CLK0, SI5351_DRIVE_2MA); si5351.output_enable(SI5351_CLK0, 1); si5351.set_freq(11998500L*100ULL , SI5351_PLL_FIXED, SI5351_CLK2); Then wire up your coax as Pete suggested here, and see if it works: ?/g/BITX20/topic/4703461 The BFO frequency is set to 11.9985 MHz. ?You probably want to adjust that somehow until you get the best audio when receiving a station. ?The value is given in Hz with a final L to tell the compiler that this is a long integer of 32 bits, not just 16 bits. ?I'm following Farhan's lead and multiplying that by 100ULL, which is a really big integer of 64 bits (an unsigned long long) but only has a value of 100. ?The extra factor of 100 is needed because the Si5351 library takes frequencies in hundredth's of one Hz just in case you care about being that precise. ?An unsigned 32 bit integer can deal with numbers up to a max of 2**32 - 1 = 4294967295, so big enough to tell the Si5351 about a frequency of up to 42.9 MHz. ?So that code could also be written as 11998500UL * 100, or as 1199850000UL. ?But then I haven't tried it. ?The "U" tells the compiler that this is an unsigned (positive) number so the bit normally used to store the sign (the + or -) can be used instead to deal with numbers twice as large. The drive_strength line above has been commented out with the two leading slashes. ?If you remove the slashes then the Si5351 will drive your channel 0 coax with a 2 ma output buffer. ?You could also try 4ma and 8ma. ? Not terribly obvious what that value should be. ?Too little drive into the demodulator and there may be intermodulation distortion. ?I doubt there's any consequences for driving at the max of 8ma, those diodes should have no trouble handling 8ma. ?But I could be wrong. ? Ideally, the 11.9985 value could be changed during the calibration sequence by pressing a button and turning the tuning knob, the BFO frequency displayed in the LCD. ?When the button is released the BFO frequency gets stored to EEPROM, where it can be read by setup() the next time the rig is powered up. ?That's left an exercise for the reader. Jerry, KE7ER On Sat, Apr 8, 2017 at 12:35 pm, Jonathan Peakall wrote:? Getting ready to convert over to using the si5351 for the BFO. I'd like to use rg-174 for both the VFO and BFO lines. The Si5351 has three outputs. The first is already routed on the PCB to the SIP header making it inconvienent to solder coax to it. Is there any reason that I can't use the two other Si5351 outputs that aren't used? That would be easy to solder to. And if so do I change the sketch like: ? |
开云体育Jerry,
>>The VFO_A and VFO_B defines are just an integer used to remember what frequency we are currently operating on Duh. I should have figured that out. Gee, ya mean like every radio with a VFOA and VFOB? As for the rest, thanks. Detailed and cogent. That gives me enough information to get into trouble! :-) >> But haven't compiled and tried it, might be sending you off with a bug or two. How dare you. When I demand help I expect it to be perfect. Next time be prepared to make a house call, bring a thumb drive with the sketch and beer. ;-) Jonathan On 4/9/2017 9:40 AM, Jerry Gaffke via Groups.Io wrote:
|
Ah, here's that bug I was warning you about. That last line of the three lines of code should end with _CLK0, ?not _CLK2. On Sun, Apr 9, 2017 at 09:40 am, Jerry Gaffke wrote:
si5351.set_freq(11998500L*100ULL , SI5351_PLL_FIXED, SI5351_CLK2); ? |
Got CLK2 going! I did have to change:
toggle quoted message
Show quoted text
si5351.set_freq(11998500L*100ULL , SI5351_PLL_FIXED, SI5351_CLK2); to si5351.set_freq(11998500L*100ULL , SI5351_CLK2); due to compiler complaints. Thanks for the help! Now off to solder on some coax and remove the xtal. Jonathan - KK6RPX On 4/9/2017 9:40 AM, Jerry Gaffke via Groups.Io wrote:
|
Good! I had borrowed that code from Farhan's original script, which compiles with the older etherkit library. All three of the lines of code I gave you should either reference CLK0 or CLK2, I gave you a mix. ?The Raduino with cabling as delivered and described in the Wireup uses CLK2 for the VFO. ?So I'd guess most people will want to use CLK0 for the BFO. On Sun, Apr 9, 2017 at 01:46 pm, Jonathan Peakall wrote:
Got CLK2 going! ? |
Jonathan,
toggle quoted message
Show quoted text
You don't need to remove the xtal. Much easier is to just remove R101 to disable the analog BFO alltogether. Or lift one leg ("thumbstone" it). If you ever want to go back to the original situation you only need to reinstall R101. 73, Allard PE1NWL On Sun, April 9, 2017 22:46, Jonathan Peakall wrote:
Got CLK2 going! I did have to change: |
开云体育Well, got the coax attached and all I
get is birdies, no RX. Not sure why. I can see the BFO clk
running. I removed the xtal and attached the coax as in here:
/g/BITX20/topic/4703461#24270 I didn't make any other changes to the code, which I may need to for LSB/USB operation. Unfortunately I ran out of time and have to take off for mt 48 hour work shift, so I won't be able to dig into it until Wednesday aft. Any suggestions as to where to look? Thanks! Jonathan
|
Alan,
toggle quoted message
Show quoted text
Drat! Impetuous youth. Or rather in my case impatient old man! I had already removed the xtal by the time I read your post or I would have just tombstoned the resistor. When I get back from work this week I'll put it back in. Thanks! Jonathan, |
开云体育I spoke too soon. Band must have been
dead when I checked it last night. Just tried and I am picking up
stations but clearly have the USB/LSB part of the code borked.
Excellent news, all I need to do is figure out the code changes.
Thanks!
|
I'm looking to make the BFO-from-DDS mod too and got the Deaf-RX-and-Birdies thing too. I was hoping you'd have some wisdom on that. :-)
to be fair, I started this before joining this group and reading the collected wisdom of many people who are WAY more experienced than I am, so I tried removing the crystal and feeding CLK1 to the input of the active section of the BFO, borrowing that idea from how the VFO is wired up. I'll try injecting the DDS BFO directly into the mixer and see if that solves my problems.? Readers from the future! If you see this email but no replies, assume that it worked. :-D |
Vince Vielhaber
How did you couple the DDS (assuming it's CLK0) to the BitX?
toggle quoted message
Show quoted text
Vince. On 09/04/2017 11:55 AM, Smitty, KR6ZY wrote:
I'm looking to make the BFO-from-DDS mod too and got the --
Michigan VHF Corp. |
How DID I? As in before? Or how am I now, as in after having read other peoples' success stories?
toggle quoted message
Show quoted text
Before: I removed the crystal and injected CLK1 (not CLK0, but they should be equivalent) into the input of the amplifiers in the BFO. That didn't work at all: deaf receiver, birdies all over the place. Now: I put a 2-pin header, ground pin in the "top" hole of the variable cap foot print that isn't being used, and the signal pin sticking out the side, soldered to a small section of the trace going into the mixer which I removed the solder mask. I removed R101 to de-power the BFO circuit, and I also removed the coupling cap between the BFO and the mixer. I probably only needed to do one of these, but whatever. Then I made a cable from RG-174 with the same 2-pin connectors that Farhhan uses on both sides. I have a stock of parts, so I just made a fresh cable, not reusing anything. There's one of these 2-pin connectors on the Raduino too. If you do this, I encourage you to keep the polarity of the connectors the same as Farhhan uses; keep ground on the same side. It just makes it easier to not have to be critical on which cable goes where. I have these mods made, but my laptop is borked so I haven't been able to reprogram the Raduino to enable the BFO yet. Working on that now. -Mark On Sep 4, 2017, at 9:09 AM, Vince Vielhaber <vev@...> wrote: |
Let's see how groups.io handles image attachments. :-)
toggle quoted message
Show quoted text
This image shows: - The now-socketed 12MHz crystal, my first attempt at injecting a DDS BFO. The additional lead length didn't seem to break anything. ? - The BFO connector I added. It's turned a little bit so it can easily reach the trace going into the mixer. That blob of solder connects to the bit of trace I scraped solder mask off of. - C106, the output coupling cap from the analog BFO, removed. - R101 (lower right corner) removed, which removes power from the BFO. - A 10k resistor (inside heat shrink) tied to C107 for the CW injection mod, as documented by amunter in his new software. (Unrelated to the DDS BFO mod.) Does this help? -Mark On Sep 4, 2017, at 9:59 AM, Smitty, KR6ZY <mark-groupsio@...> wrote: |