On 13 Sep 2023, at 22:33, ian007 via <ian007@...> wrote:

The plate voltage would still need to be like 90V-200V. Don't let the allure of $5 tube fool you, unless you already have a fairly expensive high-voltage power supply lying around looking for use.?

Martin DK3UW: It T 501 produces a substantial amount of heat wouldn't that mean big losses there ? How could that be adressed?

The new winding style for T501 in the current version of the assembly manual promises tighter coupling between the primary and secondary windings, which might put less power through the ferrite and cause less heating... I am going to try rewinding my T501 and see what effect it has.

Ted 2E0THH: How are you quantifying that, significant to what?

It's easily the brightest object in the entire QMX on my thermal camera. The thermal camera says it's about 60C, but I don't necessarily trust the thermal camera to be totally accurate. I will try digging out the thermocouple that came with my multimeter and seeing what that says.

I applied a digital probe (i.e. I poked it with my finger) and came back with a reading of "it's pretty hot, but not like, ow-hot."

At any rate the transformer is definitely hotter than the BS170s, which don't shine particularly much on the thermal camera at all. Of course this is into a perfect 50 ohms of dummy load, so mileage may vary at higher SWR.

Ted: demonstrate [with a thermal camera] that heat is NOT transferring to the bottom plane because the copper heat sink surface is in the way

Well, firstly, I'm not saying that no heat is being transferred. But it's basic physics that adding another interface in the middle is going to increase thermal impedance. Especially since I don't see any thermal compound, and those surfaces don't look lapped, so the amount of actual surface area for heat transfer is going to be pretty low.

Remember that air is an insulator. If this seems weird to you, ask yourself why we use foam to insulate things. The way that foam insulates things is by trapping a bunch of air in a way that it can't move heat via convection. Without convection, and because air sucks at conduction, the foam insulates. The only reason we even bother trying to conduct heat away via the air is because we don't have another way to do it, and we make up for it by adding more surface area to our heat sinks using fins, and adding fans to increase convection.

So what this means is that if you're not using thermal compound to fill in the gaps, or lapping the surfaces to eliminate the gaps in the first place, you're gonna have air between your surfaces. What you've done is place insulation between your heat source and your heat sink.

Are we to suppose that this copper tracing that's blocked on one side, far thinner than the added copper heat sink, and, nestled underneath another heat source, is MORE effective at removing BS170 heat than is a thicker sheet of copper added to that same original plan?

When you assembled your QMX, did you notice that soldering anything connected to ground was vastly more difficult than soldering anything else? The ground plane is an excellent conductor and spreader of heat. Yes, most of it is buried under layers of FR4 and glue. But that's more than made up for in surface area.

The thickness of the copper is not the important part. The thing that moves heat is surface area. This is why heat sinks have fins. More surface area moves more heat, a thick copper bracket with 20mm^2 of surface area is moving less heat than a thin ground plane with nearly 6000mm^2 of surface area. The thermal impedance of the ground plane to the air could be 100 times worse than from that copper bracket to the air, and still dissipate three times as much heat.

ajparent1/kb1gm: Where does the heat get radiated or conducted to?

The air inside the case, then conducted to the case itself, then conducted from there into the environment. I would expect temperatures inside the case to be several degrees above ambient. But we're not dealing with?that much heat relative to the surface area of the enclosure.

A pretty good amount of the heat in the ground plane is probably conducted to the case and the coax shield via the BNC connector.

Wow, that IS unreasonably cool and awesome!

On Sep 13, 2023, at 20:01, Martin K0MJ <oakwoodfarm@...> wrote:

I put my QMX through it's paces during a POTA activation last weekend using a windows 10 machine.? I got the same message every time I connected the QMX to the computer (either plugging it in or turning the QMX on after it was plugged in but off.? Despite the concerning message, I believe it is normal.? I just ignored the message, opened WSJTX and it worked flawlessly in digital modes.? ?Good luck.

On Sep 13, 2023, at 19:05, Howard Lamhut <Howardlam@...> wrote:

<50040412-A147-482C-8C3D-9087E267A0E4.jpeg>

Hello everyone.

I’ve been trying to get my QMX To come up as a USB device on my computer.
it comes up with the message that I copied off my screen that I attached to this post.
(USB device not recognized)
to start off I had what appears to be a fully successful build.
it receives and transmits on every band at approximately 4.6 watts out on 12 volts
I had no issue installing the firmware.
As a matter of fact I reinstall the firmware with the hopes to resolve this issue.
I am running windows 10.
i’m using a brand new cable designed for data transmission.
I tried updating the USB drivers on my computer and I have the latest drivers.

I’ve tried rebooting the computer.
i’ve tried connecting the USB A connector to different ports on the computer.
I ?touched up the USB connector on the pcb to see if that would rectify this issue.
I very carefully checked all the contacts with a 10 Power Russian Loop to make sure that my soldering was perfectly clean with no Bridges.
Short of simply trying a different computer I’m running out of ideas.?