I tried Joel's cure, it didn't work for me. However, in my random stumblings I came across the (extra) 100nF cap as supplied with my kit being used wiper-to-ground on a (10k 10t) tuning pot. My kit was supplied with 100k pot for tuning, and I figured that higher impedance could use the bypassing ... Guess what, no more clicks! and side-benefits fare more stable display, easier to follow turning of pot with displayed freq, and incredibly less static. It's almost hard for me to discern that the rig is receiving until I pinch the toroid - I'll have to check if there's any activity this evening :). Progress! Will be looking at Tx soon at this rate ... Will have to see if I can borrow (the use of?) an antenna analyzer locally. My ~40M OCF is 'by guess and by God', with a kludge-feed of 300-ohm TV ribbon and balun-plug (yeccch! That'll need an upgrade!).

Inspired by "1000 uF capacitor straight to the raduino power connector P3" approach, I tried the same (soldered 1000uF/25V directly to P3 pin header - with almost no effect.
Then added a bit of resistance directly in to the orange/+ line (see picture below)... and clicks are gone!
Resistance consisted of 3 x 10 Ohm / 0.5W in parallel (better safe then sorry) but based on voltages actually measured (see picture below) anything small enough (eg. 10 Ohm or smaller / 0.25W) would probably do as well.
For purists, an inductive choke of some kind and some value instead of resistance would be a proper way to go, but there's nothing on that board (Si5351/arduino nano) which critically requires 12V, so simple resistance is just fine.

73!
Nik, YT5AAA

I'm not a purist, so not an authority. ?But I'd prefer a resistor here over an inductor, since a few volts drop across that resistor means the LM7805 won't be heating up so much (the resistor will take some of the heat). ?And that resistor could be a lot more than 3.3ohms.

Not sure exactly what current the Raduino takes, but from Nik's figures it should be around (13.75v-13.5v)/3.3ohms = 75ma. ?The LM7805 on the Raduino wants a couple volts of headroom, let's give it 3 volts of headroom, so wants to see 5v+3v = 8v going in. ?Assuming a 12.0v supply we will have 4 volts across that new resistor. ?The resistor should be about 4v/75ma = 53ohms, and needs to dissipate 4v*75ma = 0.3 watts. ?With a 47 or 50 ohm 1/2 watt resistor in place, the LM7805 will be dissipating less than half as many watts as it did without the resistor, won't get nearly as hot. ?

Jerry, KE7ER

There's apparently no single solution that works for everybody.? I have pretty prominent clicks in my Bitx, and I haven't decided yet if they bother enough to do anything about it.
--
Darren, W9ZIM

I suppose better an inductor in series to battery wire, say30uH with10 to 12 turns on? FT37-43
Regrds
sarma
?vu3zmv

Hi Sarma,

I had at hand FT240-43 (a bit bigger than FT37-43 but the same material). With 15 wounds of CuL 0.25mm on that core, replacing resistor with the choke, clicks were ALMOST gone, but not completely while with 3x10 Ohm (cca 3.3 Ohm) clicks are removed completely.

Now, based on that specific choke should have had some 240uH which on 7MHz represents 10K reactance, far more than my 3 Ohms and yet it is still not removing clicks completely, indicating that clicks are not HF in nature. Choke having 240uH would have reactance in the range of few ohms only in the Audio frequency range. This does not necessarily mean that clicks don't have any HF component, but even if they do there seems to be an AF "signal" imposed on top of possible HF (if any) which is finding its way in to the AF amplifier through the power line.

So, following the "low cost yet functional" approach, I'd still opt for a simple resistor.

Jerry,
I wouldn't go as far as 47 Ohm b/c voltage drop I've measured was taken in "steady" state. When Arduino Nano CPU kicks in as we move the tuning pot, driving the Si chip to shift the frequency and also updating LCD, current could increase significantly above the calculated 75mA in a short burst (which is probably generating clicks in the first place). Significant capacitance (eg. 1000uF) after the choke/resistor shall provide for that increased need for a short while, but increasing resistance (unless inductive choke is used) might also lead to instability of Raduino board and we certainly don't want that, it's already working as is (and I haven't noticed overheating of 7805) even if with clicks.

Darren,
People are different, those clicks were pissing me off quite! :) So if you do decide to try it out (as it takes just a single resistor and capacitor you might have at hand), and I'm inclined to believe that capacitor does not even necessarily have to be located on P3, perhaps it could be moved closer to the resistor down the power line and still do the trick, and/or probably be smaller than 1000uF - please do let us know of the result.

73!
Nik, YT5AAA

On Wed, Jul 19, 2017 at 07:36 am, YT5AAA wrote:

So, following the "low cost yet functional" approach, I'd still opt for a simple resistor.

I'd agree. I've been working 'from the junkbox'. I junked a small PC power supply, and from that recovered a couple of 10uF/50V electro caps, plus resistors;10 ohm and 27 ohm (see footnote). I put the 10uF caps at the 7805, beside that blue pot, centre leg negative, one to each of the others. Either resistor feeding the board gave significant reduction of click, but still not totally clean. I did a bit more extraction from the PSU PCB, and this yielded a 220uF/16V. I took the 10u's off, looked up the spec sheet of the 7805 to find which pin was 'In' and put the 220u where the 10u was (they were both about the same size). Voila! No clicks, stable tuning!
My take: A decent reservoir cap and let the regulator do its job. Then a reasonable time-constant by resistor to keep the cap charged. In my case, 220uF*10e=22ms (45Hz). Bingo!Whatever floats your boat for the resistor, just leave a few volts 'overhead' for the regulator, and most of the heat will dissipate in the resistor, which could be placed anywhere in the Raduiono's DC+ line.

* I really miss that omega symbol from a convenient key! My hp keyboard has TWO backslash/pipe keys, and I run linux where backslash is a rarity (world please note:"forward slash" is tautological - "slash"? is by overwhelming use on the Internet "forward" by default! "Backslash" is a relic of Microsloth's DOS heritage. And TV Ad-men: the time taken by that useless word can be put to better use, surely?) - can I re-define the rogue? I've been thinking of disconnecting the one just left of "Z" and bridging with a copy of the rignt-hand Shift key cap), but on second thoughts an omega/Pi key would be a boon to the electronics industry. What say, hp?

I placed two 1000 uF caps at the DDS input connector with great success in reducing the tuning clicks, I opted to add a resistor in line to reduce the work of the lm7805 which was getting quite hot before I added a heat sync.
The added resistor had no effect on clicks alone.

Prior to this all my cabling is neat, short and routed nicely in the case, the main feed 13.8v feed line has a large filter cap across it also.

You can still just hear the clicks with the headphones at high volume but not at all when using the 8 ohm internal speaker, this has now made tuning a pleasant experience.

My next step will be to decouple the in and out of lm7805 just for giggles as I'm very happy with it at its current state.

73's
Matt
VK3FMLO

开云体育

I took positive lead to Raduino Board and connect to a 7808 with 1000uF +100nF on both input and output and tuning clicks disappeared?

I entered the qrp world and ordered the
ubitx v5, built properly but after running and testing on the various bands I noticed that I have clicks just as describe her.
Did you find any solution?
I use 1.2 cec firmware.
Thans

Yigal 4z4wn

Hi!

Just in case anybody reads this old thread...

I put capacitors everywhere, and I mean EVERYWHERE! 470 ?F with 100nF decoupeling everything. As well as a choke between incoming DC and outgoing. Did not get anywhere whith that. For some reason clicks where louder on USB compared to LSB. As a last resort I shielded the already short leads to the volume pot and, bliss, all clicks went away.

You learn something new every day!

Cheers!

SM6ZDM