¿ªÔÆÌåÓý

Setup uBITX for mobile and portable operations - final amps / heatsink?


 

Hey there,

I recently converted my home rig setup to a QRP SDR transceiver, and so I've been thinking about converting my uBITX for portable and mobile operations.? I know I've seen some portable setups, haven't seen many mobile.? I do expect to simplify my controls (i.e. go back to something more like the stock uBITX!), and put it in a smaller case.? To that end, my first question deals with the final amp heat sinks.

I still want to be able to use it for digimodes, and currently I have a 60mm fan that keeps the finals cool.? It seems to me than a fan is probably not ideal for mobile operations (may not have a good installation location suitable for fan intake/output) or portable operations (seems like you'd want to minimize extra ports for debris to enter).? So I'm thinking I want to use a larger heatsink (with suitable insulating kits) instead.? Is it sufficient to simply bolt the IRF510's to an aluminum case?? Or do I need an additional external heat sink?? (Or does anyone have any alternate opinions on fans for portable/mobile ops?)

Also, are there any good resources on mobile uBITX, in terms of desired mods?? I'm thinking for example of a noise blanker circuit.??

Thanks!
Rob KC4UPR


 

Hey, Rob.

I've currently got a V5 board set up as a mobile rig, separated out by the board & other daughter boards, a display head with Nextion 2.5" screen, and Vol/Channel/Function/Mic control head.?

The Board is mounted inside a former computer external drive case, large enough for a DVD/R device and its own power supply.? This allows room for the main board, a step-up buck converter for the finals, a fan already mounted inside (but run at half speed), and an additional audio amp.? I run the finals at 22V for around 30 watts out from 80 to 20 meters; the board also has 2N5109's in the pre-driver and driver positions (1 each for 2 former 3904's apiece).? There is an mounted underneath the board (as the main board's on standoffs) preceeded by a filter and followed by an . I elected to install a second Nano processor for full TFT screen utilities with the CEC software. There was significant audio interference from the TFT screen refresh, which was reduced to a satisfactory level with small torroids around the screen's output signal path between the two Nano processors. For heat-sinking:? I took two sheets of copper and ended up with an affair that sits between the finals and the outside case, and dropping below the circuit board in the case until the two sheets can fold sideways about 5mm apart as if fins under the board by a bit.? A 5V laptop cooling fan (flat profile rather than muffin-fan profile) blows across these fins while transmitting. This fan cools the heat sink and the enclosure's own fan sucks this heat out.? There are vent inlet holes on the opposite end from the case fan.? I was able to run FT8 and JS8CALL on the bench at full power & the finals (and the side of the case) would get plenty hot but would cool down rather quickly after the rig went back to RX.? Running about half output ("only" 15 watts or so), the radio could transmit digital modes for as long as I wanted into a dummy load.?

I placed the Nextion screen inside a plastic project box and trimmed around my hack-job with cork gasket material, then mounted all of that onto a device originally intended for holding a cell phone from a dash-mounted CD player. This puts the display and touch-controllability within easy view with minimal distraction from driving.? The rest of the functions, vis a vis the manual controls were placed on a mounting on the side of the car's center console at the base of the gearshift.? This panel has the mike socket, an aux socket for digital mode in/out, volume, a function button (separate from the tuning knob), and, the tuning knob.? I added a manual push-button that emulates a CW key that I call "tuning", though I suppose it could be used for CW once the zombie apocalypse begins.? There's a mic preamp/compressor in the head for voice audio and a lesser grade audio AGC for the data, since the audio drive seemed weak in the latter and I just wanted more "punch" in my voice with the former.

For cabling, I used a disused keyboard cord for the display as only 4 (shielded) wires were needed. For the "real" control head I used a DVI cable with its four each, 2-wire with shield bundles and 3 extra wires within. That's eleven separate lines, with 4 pair of them being independently shielded.? It works pretty well, actually.??

I could elaborate some more but that's the gyst of the rig itself.? I've been trying a 10-ft whip fed with a 9:1 UnUn aftrer an automatic tuner, with minimal results on TX or RX.? That part's been a disappointment.? Just today, though, I tried a shunt-fed, NMO-mounted base load mono-band 20 meter antenna I made years ago and made an easy contact with a guy at a portable campground setup over 400 miles away. ? Considering 30 watts while driving around underneath deciduous trees and power lines in greater suburbia quite satisfactory.? I made that antenna for use with an MFJ-9420 and it never disappointed. The lesson there is, be sure to use an antenna in which you've already got plenty of faith rather than experiment with something less than optimal, only to wonder if its the radio or the antenna that stinks....?? If only I could make a 40M mobile antenna that does as well.

?If I can answer any other questions or elaborate, ask.



73,

Ted
K3RTA




 

Hey Ted, lots of good stuff there!? Let me throw back a few questions:

- What's the communication between the uBITX and the remote control head?? What's the distance of that cable?

- Did you have to change any components on the board to up the final voltage to 22V?? e.g. voltage ratings for capacitors, and/or any biasing changes.

- I've got both the kit-projects BCI filter and AGC board installed.? What was the primary reason for the RF preamp (i.e. did you have a particular problem to solve for mobile?), and does it simply directly precede the BCI filter?

- Do you have any close-up photos of your heatsink arrangement and fans?

- I wasn't clear, do you have two displays?? I.e. is there a display on the uBITX main box itself, and then also a separate display head?? Or just one, i.e. the remote display?

Thanks!
-Rob KC4UPR


 

On Sun, Jan 5, 2020 at 11:18 PM, Rob French (KC4UPR) wrote:
- What's the communication between the uBITX and the remote control head?? What's the distance of that cable?
?

That's what the DVI cable is for. It seems most often used as a computer monitor cable out in the world and not hard to come by on the surplus market.? The first prototype for this was built with the original 2-line LCD display inside the control head using a 25-wire RS-232 shielded cable. There was some electronic crosstalk but it was so light that it was easily ignored.? That run was about 2.5 ft while the current line is a whopping 3', plenty enough to reach back and under the passenger seat to the radio.
?

- Did you have to change any components on the board to up the final voltage to 22V?? e.g. voltage ratings for capacitors, and/or any biasing changes.

The finals have a pair of 25V caps in the line, which are OK by a slim margin. I swapped them out for a pair of 35V caps of roughly equal value that were the same diameter & a little taller, mainly to try higher voltages. I saw no significant rise in power above 20V, though my power supply was maxed out anyway on amperage [3.5A] by the time it got to 24V.? I just called good enough, enough.?? I think I'm running about 120mA bias as a trial.

- I've got both the kit-projects BCI filter and AGC board installed.? What was the primary reason for the RF preamp (i.e. did you have a particular problem to solve for mobile?), and does it simply directly precede the BCI filter?

I found this circuit a bit deaf on compromised, electrically-stubby mobile antennas.? Not bad, mind you, but as I had good success adding the same preamp to a portable setup (see avitar) mainly because i could, it got added for good luck. One could forego that bit at first, though, given the background [cabin] noise of a mobile radio, both this and an additional audio amp helped me hear people better. The LM386 has an extra 2.2k resistor across it's input so that it avoids splatter going into a 15-amp rated amp board with its own input pot; the original vol. ctrl still acts as designed. The output to a Motorola commercial speaker is quite sufficient.?

The BCI filter comes first, and is easily wire-bridge soldered on along the short edge of this particular preamp kit board, for mounting.? The output goes to a new hole in the mainboard to connect with the AGC board on the top side.

Back to antennae:? we know the importance of a well-tuned, full-size antenna for QRP work and also of the difficulty of designing a mobile antenna that works at least a little better than a 50-ohm resistor with a length of wire trailing off of it.? My hope of working 40 meters on 30 watts to a tuned paperclip may be preposterous though there is promise on 20.? I plan to try (parked) digital modes on 40 this weekend, see how that fares as voice responses haven't been happening yet.? As mentioned, an earlier-made & known successful 20-meter antenna (on a magnet mount) showed great promise there at least.


- Do you have any close-up photos of your heatsink arrangement and fans?

No pix yet though I'll find a reason to tear the box open again, soon enough. I could draw it graphically later today....


- I wasn't clear, do you have two displays?? I.e. is there a display on the uBITX main box itself, and then also a separate display head?? Or just one, i.e. the remote display?

Just one display.? The picture of the messy work bench shows all of the sections together, more or less. Note in that shot, the mounting of the Raduino [clone] board. I? soldered a spare row of header pins onto the original pins so as to tilt the Raduino 90 degrees over on its nose to fit within the box height.. Two #8 standoffs were JB-Welded to the mainboard to offer support for the Raduino [clone] which also bears a second Nano for CEC enhancements. As there is no onboard display, the remote view screen cable connections simply attach to the Raduino former display header position. I did add a second 7805 to provide separate juice to the display.?

Next time I remove the radio for work, i.e. add a heat sensor to the fan, I'll take some more descriptive pictures. It may he a few days.? Tnx for the interest and good luck on the project.



73,

Ted
K3RTA


 

Here's a better resoution picture of the radio box.






 

Maybe try the 9 foot whip with a Silver Bullet coil? I used a tapped antenna today (I forget the brand), but it has taps for 80 thru 2 meters, and I heard a net on 40 like gangbusters. Yes, I have used the antenna before to make DX calls on the Blue Ridge Parkway. Tomorrow I may try to transmit with my new little V5!

Alex Netherton



On Sun, Jan 5, 2020 at 10:00 PM Ted via Groups.Io <k3rta=[email protected]> wrote:
Hey, Rob.

I've currently got a V5 board set up as a mobile rig, separated out by the board & other daughter boards, a display head with Nextion 2.5" screen, and Vol/Channel/Function/Mic control head.?

The Board is mounted inside a former computer external drive case, large enough for a DVD/R device and its own power supply.? This allows room for the main board, a step-up buck converter for the finals, a fan already mounted inside (but run at half speed), and an additional audio amp.? I run the finals at 22V for around 30 watts out from 80 to 20 meters; the board also has 2N5109's in the pre-driver and driver positions (1 each for 2 former 3904's apiece).? There is an mounted underneath the board (as the main board's on standoffs) preceeded by a filter and followed by an . I elected to install a second Nano processor for full TFT screen utilities with the CEC software. There was significant audio interference from the TFT screen refresh, which was reduced to a satisfactory level with small torroids around the screen's output signal path between the two Nano processors. For heat-sinking:? I took two sheets of copper and ended up with an affair that sits between the finals and the outside case, and dropping below the circuit board in the case until the two sheets can fold sideways about 5mm apart as if fins under the board by a bit.? A 5V laptop cooling fan (flat profile rather than muffin-fan profile) blows across these fins while transmitting. This fan cools the heat sink and the enclosure's own fan sucks this heat out.? There are vent inlet holes on the opposite end from the case fan.? I was able to run FT8 and JS8CALL on the bench at full power & the finals (and the side of the case) would get plenty hot but would cool down rather quickly after the rig went back to RX.? Running about half output ("only" 15 watts or so), the radio could transmit digital modes for as long as I wanted into a dummy load.?

I placed the Nextion screen inside a plastic project box and trimmed around my hack-job with cork gasket material, then mounted all of that onto a device originally intended for holding a cell phone from a dash-mounted CD player. This puts the display and touch-controllability within easy view with minimal distraction from driving.? The rest of the functions, vis a vis the manual controls were placed on a mounting on the side of the car's center console at the base of the gearshift.? This panel has the mike socket, an aux socket for digital mode in/out, volume, a function button (separate from the tuning knob), and, the tuning knob.? I added a manual push-button that emulates a CW key that I call "tuning", though I suppose it could be used for CW once the zombie apocalypse begins.? There's a mic preamp/compressor in the head for voice audio and a lesser grade audio AGC for the data, since the audio drive seemed weak in the latter and I just wanted more "punch" in my voice with the former.

For cabling, I used a disused keyboard cord for the display as only 4 (shielded) wires were needed. For the "real" control head I used a DVI cable with its four each, 2-wire with shield bundles and 3 extra wires within. That's eleven separate lines, with 4 pair of them being independently shielded.? It works pretty well, actually.??

I could elaborate some more but that's the gyst of the rig itself.? I've been trying a 10-ft whip fed with a 9:1 UnUn aftrer an automatic tuner, with minimal results on TX or RX.? That part's been a disappointment.? Just today, though, I tried a shunt-fed, NMO-mounted base load mono-band 20 meter antenna I made years ago and made an easy contact with a guy at a portable campground setup over 400 miles away. ? Considering 30 watts while driving around underneath deciduous trees and power lines in greater suburbia quite satisfactory.? I made that antenna for use with an MFJ-9420 and it never disappointed. The lesson there is, be sure to use an antenna in which you've already got plenty of faith rather than experiment with something less than optimal, only to wonder if its the radio or the antenna that stinks....?? If only I could make a 40M mobile antenna that does as well.

?If I can answer any other questions or elaborate, ask.



73,

Ted
K3RTA




--
Alex Netherton, W5ALX
Buncombe County NC, EM85ro