Good for him!?

Hi group,

?

I think I resolved this problem, years later.? ?Original problem: ATS-3A drawing way too much at 780 mA at 9V for 2.6W and over 900 mA at 12V for 4.2W.? TX current would drift down quickly and finals would heat up quickly.? Calculated efficiency about 38%.? Only observed with the 20M board.? Other boards had the typical 500mA TX current for 2.5W out and about 650mA for 4W out. Efficiency around 58%.??

?

Solution was to reduce LF2 turns from 15 to 12.?

?

I rewound all inductors, cleaned up all solder joints, and checked the value of all capacitors.? No change.? I spread out the turns on LF2 (reduce L), and the TX current went down some, but the power output stayed about the same.? Squeezed the turns and it went the other way.? So I reduced the turns on LF2 one at a time, and the TX current dropped significantly with each removed turn, but the power output stayed about the same.? I stopped at 12 turns, where I kept the 2.5W at 9V and 4W at 12V, and reduced the TX current to comparable levels to the other 3 good boards.??

?

Why did I have to do this?? Was the value of my CF3 that is in parallel with LF2 a bit too high and so I needed to compensate by reducing LF2 to keep the peak of the tank circuit where it should be???

?

I may play around with LF1 turns as well, after squeezing and spreading them to see which way it goes.??

?

Excited to finally have a working 20M board on this iconic rig after all these years (got it in 2010, 20M was always a problem for me).?

?

73

Mike N9OHW

Bumping this thread again to see if any experienced builders would be willing to help me out with this -- I'm happy to compensate you for your time.

?

Thanks again & 72,

?

Justin / NU6W

kc0ada,

assuming the crystal osc? Q8 is working properly then U3 as a bad

output and likely the BS170 as well.

?

Replace both U3 and all the BS170s..? I've fixed a few so I'm not

guessing.??

?

Ther other assumption is the assembly is correct, no shorts,

no wrong part in the wrong place, and coils wound correctly.

?

One last thing it expects an antenna with low SWR.

?

Allison

Three off? BS 170s in class? C config should give around? 3.5 W out.? Either they are not getting required drive or they have been damaged or the AC chips that are setup to drive them are damaged.? Just guesses - it's hard without having the hardware to review and test...

The hunt for the Snark continues... Snark was big in the Christopher? Robin books for? UK? reading learners. and Ditto South Seas Britishers.? The Bear and Christopher went on a hunt for one I do believe.? Poo Bear that is.

Have a look at the MST? DSB transceiver from Oz Qrz in Australia.? This company no longer exists but the manuals do.

?

you may be able to turn the? VFO/Dds you have to? ?driving a? duplicate of that? even just getting DSB out and? rx with a quality SSB portable like the PL330 from Texsun?

?

if you want the manual I have it.??

If you absolutely HAVE to have a PCB you could use the Weber? 75mtr SSB board? (which will work on 40 mtrs with minor modifications to design)? that? Far circuits sell you could use it in SSB mode? (with a bought in? 9Mhz filter from? Aliexpress) or modify it for DSB duty.? Company? Farcircuits.net in Chicago.? think the PCB is? 8 dollars US.? then input IFT is going to be hard got get. Ask FRED if he has some in his? electronics parts bins . Make abs sure what he sells has primary and secondary windings.? (for? SSB use your VFO would need IF offset and I'm not sure how hard it is to reprogam , it it might not have the? frequency range for SSB as you have it .? As for DSB generaion , it should be fine )? ?

?

If not then you can get various types of IFT? (canned transformer)? ?for Ham radio butchery from? Adam Rong at CRKITS.com? for maybe USD 2.5? Each It seems a lot but it's a seller's market.? they may not? fit the foot print for the? Far ccts board - you do the due diligence.?

?

regards,?

TEF

?

Think of it as? a little present that MR Weber** left behind to increase your? mental work and therefore you future? chance of a joyful? Mensa application.

** I forget his? first name...

I think I solved this. It is a bug with straight key mode and , I assume, the programming of the TINY45.

?

I removed the jumper that I had installed per instructions, as I only have a straight key and it is necessary anyway for accurate measurements or antenna tuning. I also tested it as closed just? during power-up, and straight key mode still exhibits the error following the opening of the switch.

?

As luck would have it, the straight key triggers dit, which is used to increase keyer speed. After setting this to something approaching infinity, a stream of dits gives a measured 1w out at 8.35v, and no heat to the BS170s. It does have an alarming current draw, approaching 2A.

Is resolution? of this anywhere on the horizon now that the Weekend has freed up? "spare"? TIME? ??

?

i'm wondering? - specific? component fault or? just a crummy RF design to start with ??

?

?

As the man said - there are only so many hours you can spend polishing a turd.?

?

?Possibly accept that? the? Vfo works and the rest no so much.??
Clearly there is a hard bias getting on to at least one gate.??

?

FYI? Farcircuits.net? ?Mr? ? FRED of? ? Chicago sells a? SSB? Tcvr board for about seven dollars? that is designed for? 75 m designed around? a S WEber brain wave.?

If you can get a VFO going and?a 9 Mhz? ladder filter strip? (? ten dollars? from? ?Hamradioshop? store on? the? ALEXP , but you do need to get two off? 9 Mhz xtals from somewhere else ie another vendor? for? carrier injection Osc use)

Then? ?you can build a 5 W or? 2 W on 20 mtr? SSB rig which can also? produce? CW if you inject a? pure tone into the microphone port .? You'd have to pull up the Survivor? radio schematics from Pacific Antenna? ?CAlif? to see how WEBER? modified the RF? out switching because what? he did on his first pass would make th e FRED version very deaf .? At least that makes some use of the VFO component as long as it has the ranges to suit? the desire? Rx band WITH IF Offset?

(I believe there is enough copper on the pcb to make the required change - adding some BS 170? Tx/Rx switches.)?

As far as I can tell, the SA612A is used during receive, but has no role during transmit. Receive is working OK, right?

Ok, I am about to pull the NXP SA612A, or at least lift pin 8. I wonder if it has failed or is not what it says it is. Please tell me know if this is wrong!

Here is the equivalent schematic:


Based on this,? I would expect the following resistances:
7 -> 3 = 25k
1 -> 3 = 1.5k
2 -> 3 = 1.5k
6 -> 4 = 19.5k
6 -> 5 = 19.5k

6 -> 8 = 18k

Measured:
7 -> 3 = 23.7k
1 -> 3 = grows to 1.47M one way, switch probes and it grows to >2M
2 -> 3 = grows to 1.47M one way, switch probes and it grows to >2M
6 -> 4 = 18.1k
6 -> 5 = 18.1k
6 -> 8 = 16.6k

Thanks for the suggestions on potential improvements, Timothy. I will save these schematics and notes for review later - my goal is getting the initial design to function (or really understand why it is failing) and go from there.

Seems the schematic for the? power RF osc etc of the Bayou jumper did not attach- here it is. Better used? as an amp being fed from a 5mW VFO.?

If all else fails in terms of the Tx fro the? DC transceiver? .. here is the RF Deck? of the? Bayou Jumper .? This is single band but the extension?

to three bands is obvious.?

if the Xtal is removed and the LP rf from the DDS unit? which is working? is fed in to the base of the? 2n7000 via a 100 nF capacitor

you have about? 4 W out at? 12 V (based on design values when it? is used with Quartz )? and about? 6W out? at? 15 V.? ?If running at? higher voltage add 9V zener stab to the low power? Mosfet on supply side.?

There is no PCB layout available for this .? The? design From? 4state Qrp group in the US.

?

The RX is? supergenny and work of the devil so don't go near it.

Block diagram is completely AWOL lfrom the? referenced URL