Hi all,
One of the best threads from the old forum was the "Whats on your Workbench" thread.
It was great to see what people were up to and contained a lot of useful information for the home brewer.
So.... lets start a new one
Like most hams I have n+1 projects on the go, currently I've been fitting an ATU bypass to my HF amp so I can insert a 1kW ATU between its output and its inbuilt antenna switch. See pics
There's a couple of forlorn pcb's sitting there waiting for something more to happen along with a couple of other half completed things underneath the mess plus a roundtuit list a mile long.
My 'software' workbench has had a project mostly-finished and triggered by the recent rain here in VK4. It was a thought back last year but the rain made me sit in the recliner and put it into practice:
The software does a 'grab' of the selected BOM rain radar and saves the image to local file once every 6 minutes - with a few provisos. You can then open the 'save folder' and view all images with an application that allows rapid scrolling between image files (I use Imageglass) and you can see the movement of the rain over time. (You slowly lose it if you directly view any of the BOM radar sites.)
All Aus BOM radar site details can be accessed and the selected details are saved for a subsequent session.
The downloaded GIF image size must be greater than the user-set threshold size to be saved to file, anything too small is ignored. You set the reference threshold only when there is no rain evident on the radar of interest...
I have a copy running on a low performance laptop that runs 24x7 and it provides a great graphical history of when & where the rain has fallen.
No current plans to make it into an article on my web site but if there is enough interest by others, maybe.
This Forum is only going to be as interesting as the posts it contains. If you have a comment or question, post it as it may trigger or answer the query in someone else's mind.
Very nice Doug.
Dunno why they don't keep a longer history in their loop that is a user setting. Or even the ability to look historically.
Mind you those are things that are probably part of their paid service.
Next project is to build one of the Chinese MRF300 500w 2m amplifiers that can be found everywhere.
Documentation is sparse at best but I've managed to find various schematics, parts lists and some of the original NXP data it is based on.
The intent is to document the whole thing and make that information available to all
So far I've assembled the main components on the PCB.
Next thing to do is to sweep the filter to see if its any good.
(31-12-2021, 04:58 PM)VK2XAX Wrote: The 2m amp is taking shape, sweeping the filter shows it to have negligible insertion loss, see pic.
The spec analyser screen shows -47dB at 2F but looks to have about 2dB loss at 144 ?
Now the maths is that if 2dB is accurate as insertion loss in the passband means that at a 500W level into the filter, there is an output of power of around 316 watts. That could mean that 180-odd watts is going to be dissipated in the output filter !!!!! On SSB that might be an acceptable heating level but digital mode could mean burn-up.
How accurately have you measured the actual filter loss at 144 ?? What component types did you use ??
This Forum is only going to be as interesting as the posts it contains. If you have a comment or question, post it as it may trigger or answer the query in someone else's mind.
(31-12-2021, 06:23 PM)VK4ADC Wrote: The spec analyser screen shows -47dB at 2F but looks to have about 2dB loss at 144 ?
Hi Doug,
The SA screen doesn't show the loss accurately when you have a large range to look at. The measured loss was no more than 0.2dB for the filter, certainly not 2dB! - that would be bad
The coils are hand wound 1.2mm wire and the caps are ATC 100 types as supplied with the kit.
The intent is to assemble and test this kit with parts supplied to see if actually works as advertised. After that I might make some mods to improve its behavior if needed.
i.e. the bias circuit has no temperature compensation but that can be achieved by changing out the 10k resistor in the bias voltage divider for a 10k NTC that I'll place in close contact with the heat sink.
Phewww ! (Wipes sweat from forehead..)
The 0.2dB figure is a heap better but will still be interested to (eventually) hear how hot your LPF inductors are immediately after RF is removed after running at/near full power.
When you get to measuring actual amp output 2F and 3F product levels with the SA, set your 2M reference level then add a series 2M L-C notch circuit across the output of the power sampler/coupler (use a BNC or N Tee with a short cable length to the notch) to attenuate the base signal level before you look at the 2F and 3F levels.
Otherwise the SA is likely to be in signal overload from the 2M fundamental, particularly if you change the SA input attenuator settings, and your readings will be largely invalid.
I use a shaft-control air-variable C and a few turns of tinned copper wire in a small diecase box and my notch covers from about 105 to 190 MHz.
After setting the reference level, simply add the notch and watch the 144MHz signal component as you tune for a minimum. Usually about a 20-25dB drop depending on construction.
The LC notch will be largely 'invisible' to 2F and 3F products.
This Forum is only going to be as interesting as the posts it contains. If you have a comment or question, post it as it may trigger or answer the query in someone else's mind.
Hi Doug,
Great idea! - I'll also be using a different SA for the final tests. The SA in the HP8935 test set is OK for general work but to look at things in detail, and with a substantially better dynamic range, I'll be using the HP8566B along with appropriate attenuators and directional coupler.
cheers
Tim
Hi all,
A little bit more progress tonight, 1 minute key down to see what gets hot.
Replaced the unknown blue coax balun that got hot with with RG400. That gets mildly warm now.
Added a 10pf ATC100B across input to balun ( I saw this on other designs and wondered what effect it might have, it increased the gain ) I experimented with other values either side of 10pF but that value is optimal.
Output power is now 56.6dBm aka 457w @ 50v and 12.5A so that computes to 73% efficiency
Additional 1 minute key down to see what else gets hot and found that the 12.5ohm coax used for the impedance transformer gets quite warm, so I'll replace that with something beefier.
The 1.3mm wire in the LPF isn't going to cut it either, they get too hot to touch so I'll upgrade those to 2mm enameled wire.
![[Image: smile.png]](https://www.ahrdf.net/forum/images/icons/smile.png)
