开云体育

Hi David,
With the ATU in bypass, does the meter show a PWR/SWR reading when you transmit?

If not, then check that injecting a DC voltage on the Stockton bridge diode D2 (Cathode) shows a reading on the display. With around 2.7V on the cathode, the meter should read about 10W.
If this doesn't work, then check the potential divider R4 and R21, there should be 0.9V on the PIC with 2.7V at the D2 (cathode).
73, Dave

Hi David.

So today finally got round to applying RF into the input and a 50 watts Dummy Load on the Ant Output.

With 10 watts out of my known transceiver absolutely Zilch nothing and no relays chattering as there should be.? So done some continuity? checked all of the input through the transformer to the L1 infact right through all of the coils and continuity right through.

Also checked continuity through from Ant Output to the diodes and again ok.? So I am totally baffled again.? Could you throw any light onto the problem as I don't know what to do next.? I wound 10 turns on one side of the Balun and 10 turns on the other side.? Started the windings at the bottom with thin PTFE wire.? And One covered PTFE wire through the middle once and terminated at each side.? Cannot see what I could do wrong there.?

Maybe you have a quick test to check all is well BUT clearly is not.

I've also checked the polarity of the diodes and checked they are working.

WHEN I check from input to Output I get a short because of the transformer windings and imagine you obviously get the same.

?

I was waiting for all of the relays to start singing to me but not a thing.

?

73s David? ?G4RVH

Good Morning David? ?G8PTN

Thanks for you kind words. But it sounds like you and I are alike in this radioham hobby,? a labour of love.? Also have the MFJ258 for very quick checking. I have thought of buying? the Tiny SA but tend to use my homebrew QS1R Quick Silver receiver designed by Phil Covington and Zoom right into the signal looking for nasty spurs.? Cathy developed the Software for the QS1R

and also I use her NanoVna software which is fabulous because if you point the mouse onto a desired part of the screen it will give you a Match result L & C.? I do very little operating with a Mike or key these days but have experimented now constantly 24/7 with WSPR to test my aerial systems.? The very best I've ever achieved at my location is with the setup I have now. A 160 foot Doublette fed with homebrew 450 ohms open wire feeders 30 foot long. The results have been staggering.? Been heard in VK literally thousands of times this year with 100 milli-watts and down to South Antarctica but the best was a VK on just 1 milli-watt.? But the most interesting thing is actually monitoring where the aerial is firing and actually ditto with Roy Lewellen Software EZNEC.

ROY has given his Software EZNEC PRO VERSION to all radiohams as a retirement present to us all so grab a free copy.

Kindest Regards to you and your XYL of course.? Great speaking to you David.? ?73s? David? ?G4RVH

Hi David,
I am more than happy to help out where possible and hopefully the ATU-100 will work when you put RF through it.
Yes, the enclosures can also take some time. I tend to use the old school method of drilling holes, and then cutting/filing manually.
For projects that don’t need screening, I tend to use ABS enclosures, and I have also printed a few with my 3D printer.
Regarding impedance measurement test gear, I started out with a good old MFJ-259, then an FA-VA5 Analyser from SDR-Kits, and of course the NanoVNA.

The TinySA is another great bit of test equipment to have on hand.
To be honest, I find the construction and experimental side of things is the best part, there is always something to learn and/or understand in more detail.
Anyway, keep up the great work and best wishes to you and your XYL.
73, Dave

Hi David? G8PTN

You have been very useful guiding me through my difficulties and most grateful of your valuable help.

My ATU was not a kit as you know so had to source all of the parts.? The Chip 16F1938 wasn't that easy because the first one didn't fire up so had to take it off and solder another one on the pcb.

Even the aluminium extrusion box was a swine to cut with a tiny round dremel saw.? But at least I can now see what's going on in the box.

Also by checking with a VN?A your sure all the joints are soldered correctly and the relays all functional.??

Before I bought the NanoVna I had a Polyscope for sweeping Bandpass filters.? It weighed a ton and was massive.? Did you use one.? A beast.

But of course time has marched on and the NanoVNA is a lab in a box.

Oh I did build the Paul Kiciacs N2PK which got me by sweeping filters in real time with the faster ADC

73s? David? ?G4RVH

Hi David? G8PTN.

Sim Smith is good to get the matching to be nearer enough. I put my NanoVna onto the ATU tonight actually just to make sure that I could actually see the L's & C's altering the display on my Laptop after Calibration of course OPEN Short Load isolation plus thru the Vna.? Then put a 50 ohms dummy load onto a T BNC on the output of the ATU.? As you said its tickly unless you know what values to poke into the ATU.? It took me 20 minutes to get a match.??

73s David? ?G4RVH

Hi Dave? G8PTN

Attached video of both banks of Relays.

Watch the Bottom Middle S8 led go out switching relay banks from L to C.

73s Dave? ?G4RVH

Hi David,
Excellent news that all is now working okay.
It’s probably easier just to go for an RF test with 5 ~ 10W.
I would initially put the ATU in bypass with a 50? dummy load and check that the PWR and SWR readings are as expected.
The auto tuning relies on the PWR and SWR reading for its tuning algorithm.

I also did some manual tune tests with 27?, 50? and 390? using my NanoVNA, but it can be very time consuming, and you need to get close to the target values.

I actually used simulation results to find the theoretical matching values.
73, Dave

I Dave? G8PTN

Just had a thought.? I will test that the Inductance and Capacitance will tune the 50 ohms at a frequency with my NanoVNA on the Smiths Chart.? One stage closer.??

73s Dave? ?G4RVH

?

Hi Dave? G8PTN.?

So this morning took the driver Mosfet off the board that drives S8 Relay. Replaced the Fet and in Test Mode all working as it should switching between the Capacitor bank and the Inductor bank.? And of course S8 switching correctly.

So I am finally down to the last phase of testing the Bridge out.

Have you a very easy way to test the whole unit out rather than squirt 10 watts of RF into the input.

I've also checked all of the relays switching over and all ok.

All the Outputs on the 16F1938 are short circuit protected apparently so been very lucky. 73s Dave? G4RVH?

Hi Rob,
The ATU-100 does work without modifications.
The power rating is somewhat optimistic since the component types/ratings are not necessarily as specified by the original designer.
It’s possible to change various parameters that are stored in EEPROM memory, but a specific programming tool is required. The programmer is also required for any firmware updates.
There are several different versions of the ATU-100, some with all buttons fitted (Tune, Auto, Bypass) some with batteries and in pre-built or kit form.
It’s probably a good idea to have a read of the user manual and look at some of the on-line reviews to see if the ATU-100 meets your requirements.
The manual is available from the GitHub link below.

73, Dave

Guys.

Was looking at purchasing one of these units, but with all these mods everyone are doing, is it worth purchasing?

Cheers

Get BlueMail for Android

Hi David,
If there is a short between the "Gate" and "Source" of the MOSFET, it may have damaged the PIC, since it would have been driving a logic ‘1’, which is now shorted to ground.
It also seems strange that two boards would have the same fault.

?

With a 10:1 turns ratio on the Stockton bridge, the calculated power reading based on the equations used in the PIC firmware are shown below.

When I first built the ATU-100, with a 10:1 ratio, I tested up to 9V at the Stockton bridge and the readings were as expected.

If the voltage injected on the “RVS” input is greater than that for the “FWD” input, the firmware will show an SWR of 9.99.

?

73, Dave

?

FIRMWARE CALCULATED POWER (1)? ? Based on voltage at Stockton bridge.

?

?

FIRMWARE CALCULATED POWER (2) ? ?Based on voltage at PIC.

?

Hi Dave? G8PTN.

YES the ATU is doing exactly the same as your photos so obviously performing correctly.? Hopefully this week after changing the S8 Relay FET all should be OK.

Thinking ahead I might have trouble with the Shottky Diodes I've used.

I did manage to poke 2 volts DC into one of the input ports and a reading came onto the screen something like 20 watts out? 99%? bla black. But the other Reflected port showed nothing

probably because it's looking for a Negative volts.? Am I correct ?.? I only did it for a very quick check.

I knew that you cannot put more than say 4 volts MAX into the Input Ports.

73s Dave? ?G4RVH?

Hi Dave? ?G8PTN? the reason I put your callsign is so readers can distinguish between you and I.

Your so kind to take the time out to help me.? It's so much appreciated believe me.

Anyhow it's looking like good news because I've just checked the continuity between a good relays that is switching.? With the ATU switched off its obviously showing a short between the Negative rail to one of the FETS drain.? BUT no short on either the Source or Gate.

BUT on S8 Relay I am getting a short between the Drain of the FET and the Relay which is obviously correct but also getting a short between Gate and Drain and Source so must have blown the FET.? Things are looking up.??

73s? Dave? ?G4RVH

Hi Dave? G8PTN

Just a few photos of the bottom of the ATU box to help me servicing in the future.? I will replace S8 FET this week and hope this cures the problem of S8 Relay being permently being switched on.

73s Dave? G4RVH

Hi David,
Unfortunately, I can’t recheck the behaviour on my ATU-100 at the moment since I modified the firmware, but I am sure when I first built it, everything worked as expected.

To summarise.
- On power-up the S8 relay should be “ON”, capacitor bank on output side.

- Put the tuner into manual test mode by holding “BYPASS” and “AUTO” button during power up.
A long press of the “TUNE” button should toggle between adjusting L or C manually. This is indicated by “l” or “c” on the display.
A short press of the “TUNE” button should toggle “S8” and should show the C value above or below the L value on the display.
Note: The “Tune” button short press time needs to be greater than 50ms (de-bounce protection), but less than 250ms, otherwise it will be treated as a long press.

?

I don’t think it’s a good idea to add a switch, since when auto tuning, it will also select between input and output side to get the best SWR.

I found a few screen images (see attached)
73, Dave

?

?

Ditto Dave I meant to say to you that hope you are well also my friend.

I suppose if all else fails and that transistor that feeds S8 refuses to switch off THEN I could fit a switch on the front panel to Switch S8 off physically.? ?????

?73s? Dave? G4RVH

Good Morning Dave? G8PTN.? Thanks ever so much for trying to help me out.? Luckily I have two of these PCBs and they both have the same fault whereby you switch the unit on just the main switch. Then if you press the Green press switch in slightly longer a little c appears and if you press the Tune button again the c goes out but unfortunately the S8 Relay never switches off at all whatever I do so imagine having tested 2 boards now that I definitely have a fault somewhere.? I've been going round in circles.

This morning I tested a brand new relay so definitely know the Normally Open and Normally Closed contacts in readiness to test all 15 relay contacts out just to make sure that they are all switching over.? BUT until I can be able to switch off Relay S8 I am struggling.

Many thanks for the detailed explanation? ?

73s? Dave? ?G4RVH

Hi David,
I hope you are well.

This is the correct behaviour. On power up “S8” defaults to “ON”, which means the capacitor bank is connected to the RF output side.

When you are in the manual tune test mode, short press the “Tune” button to toggle between capacitor bank on input or output side of the network. This is covered in the manual (see attached).
When the capacitor bank is on the input side (S8 OFF), the C value is displayed above the L value. If the capacitor bank is on the output side (S8 ON), the C value is displayed below the L value.

73, Dave

?