When I finished building the kit I had receiver problems, all traced to bad/cold solder joints. The transmitter appeared to be working fine, 5 watts out with 15 VDC supply. Then while further testing, I found a bad coax jumper which caused me to transmit briefly into a high SWR. Since then power has been low (2.7 watts) and high current draw on transmit (700 ma). Receiver is superb so I'm assuming T1 is correct. Not having real test gear, I shotgunned it by replacing the 3 finals and Q6. Nothing changed. Can anyone offer suggestions as to what to check next?? Are there any other component failures that could cause it without affecting the receiver?

Thanks,
Jim - K9DD

I built the kit a few weeks ago and want to use it with GW3UEP's 100W PA on 630m. The last post I can find relating to CW is #18874. Ken G4APB notes that TX CW works as long as SHAPING is set to 00. An earlier posts notes that the Potentiometer tune function does not work.

Am I to assume that in CW there is no waveform shaping and I should implement shaping elsewhere. Perhaps as GW3UEP shows in his 100W PA keying circuit.

Brian VE7JKZ

Tim,

Don't be silly... of course its macro hurts.

Allison

Justin,

I can see wanting or needing the bandwidth.? FYI the bs170 will not make it to 300mhz
and even 6pf at 100mhz is getting to 200 ohms.

If cross talk is the concern the easiest way is multiple 5351s.? IF driving levels are the problem
then maybe a fast video driver or MMIC.

The 5351 can drive fairly low impedance but the cross talk to other channels goings up.
ITs the trade for that level of capability and low cost.

Allison

I have considered the single gate 74 series inverters, so far the fastest one I can find only goes to 100MHz. I'd like to support the entire range of the Si5351a. so as close to 300Mhz as can be would be nice.?

-Justin N2TOH?

开云体育

On 14/09/2018 12:11 PM, ajparent1/KB1GMX wrote:

[snip]

It the BS170 has a input C of 60pf at 3mhz that's about 800 ohms not
counting Miller feedback and at 30mhz your down to 85 ohms.

millihertz ???

regards

Tim

-- 
VK2XAX : QF56 : ITU59 : CQ30

Hans, thank you very much for the explanation of CLK1 not enabled when receiver is muted.? This transceiver kit has certainly been an excellent learning experience!

73 John KF6QPC

On Thu, Sep 13, 2018 at 07:11 PM, ajparent1/KB1GMX wrote:
Cant say as the input capacitance of that type of mosfet is not that high.

should be:

Can't say as the input impedance for that type of mosfet is not that high.

Allison

Cant say as the input capacitance of that type of mosfet is not that high.

Keep in mind at DC its infinite.? At RF its current flowing into and out of
(source and sink) the gate capacitor.? You want it very small.?

It the BS170 has a input C of 60pf at 3mhz that's about 800 ohms not
counting Miller feedback and at 30mhz your down to 85 ohms.

Consider a 74AUP04 (single gate cmos inverter) with a 4-7pf input C.

Allison

Allison,?

the end use is driving an SBL 1 mixer, the reason I ask about the input impedance is I want to minimize the loading on the Si5351a synth chip.?

-Justin N2TOH

Like most fets both junction and mos the input impedance is far higher than?
50 ohms and also reactive. To get a fet to match to a DBM output requires
a feedback circuit and often resistive loading.? The presumption is the DBM
is outputting to the mosfet.

The output can be low enough to match 50 ohms for driving.? This assumes
the mosfet is driving the DBM.? However the standing current has to be fairly high.

What confuses me is you ask about the fet input impedance and talk about driving a DBM.
Its two different things.

Allison

On 01/09/2018 18:49, Alan G4ZFQ wrote:

Did my first one this afternoon using an Arduino to program it on the board.? The information for the settings including the fuses are here?/g/QRPLabs/files/AVR%20Programming and reading the aVRDundess1.pdf file.

Len,
I don't know why but the fuse settings do not seem to agree with Hans /g/QRPLabs/message/24845

73 Alan G4ZFQ

On Thu, Aug 30, 2018 at 09:06 PM, Len GM0ONX wrote:

As a learning exercise I want to load on to a spare ATMega328p the version 12a hex code using AVRDudess.? I've found some info on here including the version 12a hex code but not the version 12a eep file refered to in AVRDudess only the older versions.? Any pointers as it been years since I've programmed an AVR and it all changed.

Len GM0ONX

They say bride comes before the fall.? I blew my AVR by failing to remove the link between the PA and +5V and applying 12V. Unfortunately the replacement got stuck in the Turkish mail system for ages so I thought I program my own.

Everything looked fine but it didn’t see the Si5351. Long story short, the proper AVR turned up and it works perfectly with Han’s version of 3.12a of the AVT but not the one I programmed. I programmed it using an Arduino ISP sketch, AVRDudess with the recommended setting L 0xF7 H 0xDC E 0x07.? Everything in the menues looks fine but the clock runs really slow. Any thoughts?

Hi, Hans?

if I may ask what is the input impedance of the BS170 based amplifier? I'm interested in driving some Mini Circuits SBL 1 mixers and had hoped that the BS170 single device amps would fit the bill.?

-Justin N2TOH??

Hello Chris and Alan, after hooking up my oscilloscope, I noticed both CLK0 and CLK1 are active.? The only? time CLK1 would be 0 is after placing the QCX in menu 9 Test Equipment | 9.2 RF Power and depress the key.? If I move up the menu back to 9 Test Equipment, CLK1 is active once again.? I suppose because CLK1 is missing only during test 9.2 plus key depression, this observation now becomes a mute point.

I want to thank you both for taking the time responding to my question/concern.

73 John KF6QPC?

Hello Alan,

I think this is the info he needs from a quick search of my folder:

"This has come around before, but here's a reminder.

The part you want is the Si5351A-B. It comes in a 10-MSOP package. You
can buy it either on a cut tape (-GTR) or bulk from a tube (-GT)
according to taste; same chip either way. You want to avoid the
variants that have a modified I2C address such as the Si5351A-B04486;
the QCX firmware expects the chip to use the standard I2C address,
0x60. You also want to avoid the Si5351B-B or Si5351C-B, which come in
larger square packages.

Digikey sells them for 92 cents in quantity one. They're a bit more
expensive at Mouser: $1.15. (But it might be worth paying the higher
price to Mouser if you're already ordering from them for other
reasons.) Non-US distributors also sell them. "

Thursday, September 13, 2018

--
Best regards, Chris Wilson (2E0ILY)

As mentioned previously, after building the QCX, the voltages of both clocks WERE the same

John,

Yes, but bad joints/bridges can show afterwards. Have you checked for grounding or with the scope yet?

and for some reason CLK1 is now 0v.? As for the SI5351A_MSOP, it is my understanding that this IC may be a pre-programmed component.

The one used is not pre-programmed, if you look back in the posts (search) the exact replacement has been quoted in the past. Maybe someone will know and save you the trouble.

73 Alan G4ZFQ

Hello Alan,

Thank you for your reply.? As mentioned previously, after building the QCX, the voltages of both clocks WERE the same and for some reason CLK1 is now 0v.? As for the SI5351A_MSOP, it is my understanding that this IC may be a pre-programmed component.? I need to know if QCX's IC1 is pre-programmed or not.? IF programmed, how do I go about getting one?? Otherwise, I can order a non-programmed one from Digi-Key.? The IC4 appears to come in a couple shapes/sizes.? I don't think I can order the correct one (rectangular) within the States.? I know Digi-Key only offers the "square" one.

BTW, I did reset IC2 but that did not improve this situation.

BUT CLK1 was 0!? I haven't yet hooked up my scope but I am assuming something is holding down CLK1 or I have a failed IC.? I notice on the schematics (Rev 1.10) that CLK1 is produced by IC1 p9 (SI5351A_MSOP) and CLK1 is shared with IC4 p2.? My worst-case scenario is a faulty IC1and/or IC4.? If IC1 requires replacement, is this a programed IC?

John,

A few faulty components have been reported but faulty new components are rare.
More often faulty components are due to a constructor's mistake.
Most faults found are due to bad solder joints or solder bridges.
Bad soldering on the Si5351 has been reported a few times. A very careful look at all joints with magnification is essential.

The components are available, yes, if you decide you need something make sure it is the correct type.

73 Alan G4ZFQ