600 watt HF Linear Amplifier Project

Discussion in 'Homebrew and Kit Projects' started by KD8UYQ, Feb 9, 2017.

ad: L-HROutlet
ad: l-rl
ad: Subscribe
ad: L-rfparts
ad: l-assoc
ad: l-gcopper
  1. KD8UYQ

    KD8UYQ Ham Member QRZ Page

    Hello All,

    I recently realized that I forgot the switch to manually apply the bias voltage to the PA board if I don't have the linear connected to my Icom 7300 so one was added. It is the square black button to the right of voltage display.

    Bias switch on cabinet.JPG
    When the push button is out the bias relay applies the voltage to the PA board upon activation of the transceiver's PTT switch. Push the button in and the bias voltage is on the power amp board all of the time.

    Another thing that has bugging me was what to do with the 1:1 balun on the output of the PA board's transmission line transformer. I didn't want it just floating around inside of the cabinet. Scratching my head for a while I came up with this solution.

    1-1 balun standoff.JPG

    I used a control surface clevis for remote control airplanes as a pivoting standoff. The nylon and wire standoff is attached to the coax cable between the LPF and trans-match boards. When the bottom panel is in place the LPF coax cable holds the balun away from that board and in space above the trans-match board below it.

    1-1 balun standoff bottom open.JPG

    When the lower panel is laid open the balun keeps its position in relation to the LPF board coax cable and is still held away from the low pass filter
    toroids below it. Once those two mini projects were completed all that was left is to wire up the DB-9 connector for the boards on the bottom panel and that is were I am now.

    wiring  bottom panel DB-9 connectors.JPG

    Ripley
     
    AF6LJ likes this.
  2. N1OOQ

    N1OOQ Ham Member QRZ Page

    How well do all of your toroids hold up to the power levels? Any of them get hot?
     
    AF6LJ likes this.
  3. KD8UYQ

    KD8UYQ Ham Member QRZ Page

    No sir. The toroids run cool as a cucumber. The Low Pass Filter Diplexer board in particular, is rate for 1.5 kw so it doesn't even breath heavy.

    Ripley
     
    AF6LJ likes this.
  4. KD8UYQ

    KD8UYQ Ham Member QRZ Page

    Hi Again,

    Ninety percent of the cable plant has been completed. It is taking a while because I have been trying to keep the wiring and cable lengths short, account for the hinged bottom panel, and pay attention to lead dress. I want the wiring to look neat and be easy to trace what wire goes where.

    bottom panel wiring.JPG
    wiring left side.JPG
    The wiring in and around the front panel favors shortness of wire length over lead dress because on the mock up board the displays were the most effected by RFI. This view is from the bottom of the cabinet with the linear amp upside down.

    Wiring front panel.JPG

    The Mu metal RF shield will cover almost all of the wiring around the front displays.

    The vast majority of circuit boards are wired via connectors of some sort or another including the little LED mini circuit board for the high SWR and input over drive warning lights.

    Warning LED board.JPG

    Should I ever have to do anything on this little guy all that I have to do is unplug the two cables and release the hook and loop fastener that holds this assembly to the front panel.

    Warning LED board mounted.JPG

    Now that the power and SWR LCD has the supply voltage run to it the first thing that I noticed were the blue LEDs that indicate which band is selected are way brighter than the LCD display. The LEDs looked like a flash light that was shown in your face while the LCD looked more like a night light behind that Plexiglas panel. To fix the brightness differential I reworked the selected band LED circuit board to incorporate a trimmer pot that adjusts the brightness of the blue LEDs. These pictures were taken after the brightness adjustment trimmer was added.

    Linear front panel power on.JPG

    LED brightness control.JPG

    Linear front panel power on pic 2.JPG


    Sigh.........

    Now the Power On LED looks to bright compared to the others

    Ripley
     
    AF6LJ likes this.
  5. KD8UYQ

    KD8UYQ Ham Member QRZ Page

    Hello All,

    Life and family have taken center stage the last few days but with Easter behind us I have done a few more things on the check list to finish this project up. The heat sink now has a cooling fan permanently mounted to it.

    Heatsink fan.JPG

    All of the scratches on the cabinet have been painted, the power LED was dimmed down a bit so that it's illumination is on par with the other LEDs. And I started to add some labels.

    Lettering on front panel.JPG
    I came across this vintage decal booklet at the Radio Museum the other day and gave some of them a go on my cabinet. After all these years they still applied well but to avoid any silvering I dabbed on some decal setting solution. Once they have set for a day semigloss lacquer is brushed over the decals to help protect them from wear.

    Water slide decals.JPG
    In the mean time I am having two more panels cut from real smoked glass this time. When they are done the panels will be taken to a glass engraver to etch the band indicators, HSWR, and Over Drive warning nomenclature, among others into the glass so that the LEDs will illuminate the lettering. I not sure if the best look will be with the etching done on the back or the front of the glass so I am having it done both ways.

    Up next are the feet for the cabinet, finish up the wiring to the PA board, make a removable filter, and last but not least fire this thing up to see if it is more than a pretty black brick.

    Ripley
     
    AF6LJ and N1OOQ like this.
  6. KD8UYQ

    KD8UYQ Ham Member QRZ Page

    Hello again,

    Time for an update. I made feet for the cabinet from 1-1/2" aluminum bar stock that were turned on my mini lathe to make them look bright and shinny. Then they were cut into 1/2" thick hockey pucks like so.

    Aluminum feet.JPG

    The center of each foot was drilled and tapped to take a 8-32 machine screw. The feet are attached via Phillips head machine screws from the inside of the cabinet, through the holes in the bottom plate, and into the threaded holes in each foot.

    Feet on Cabinet.JPG

    Finally 1-1/2"self adhesive black pads were a affixed to the bottom of each foot to protect the counter top.
    Cabinet sitting on feet.JPG

    Those black pads are not shown in the picture above but you get the idea. Now that I had a space under the cabinet to draw in air for the exhaust fan on the top of cabinet it was time to finish up the wiring and test the amplifier's boards in there new home.

    Icom and Linear Amp test.JPG

    I am happy to report that everything works with only a few minor issues. In the picture above the linear is in automatic band selection mode (Orange LED to the right of the band selector knob) and the Icom has told the linear to select the Low Pass Filter for 75 meters (Blue LED at the 10:00 position around the the band selector knob). The linear is putting out 500 watts into the dummy load and the displays are showing a current draw of 11.7 amps while the supply voltage is showing 46.7 volts. The temperature of the copper heat spreader is 36.8 degrees C.

    The first issue is I don't believe the current display. The reason that I say that is when everything was mounted on the mock up board the current display would show the 600 mA idle current when the bias button was in the manual position. Now that all of the boards are mounted and grounded to the metal cabinet the current display shows 0.0 mA under the same condition. These Chinese current meters measure current from the negative side of the power source so I suspect that I have another path to ground somewhere and the current display is only giving me a partial picture of the current draw in the system. Something to explore further but not a big deal.

    Icom and Linear Amp test 0 current.JPG

    You will notice that the first picture of the linear amp under test, the green LCD display is showing a power output of 484 watts and a SWR of 1.43. The displayed numbers jumps around a good bit at 80 meters but calm down and are more steady as you go higher in frequency so I suspect that RF is getting in to the display. I am going to add more shielding around it because, at the moment, only half of that board is shielded. Also the two Forward and Reflected power wires that come from the transmatch board and up to this display to are not shielded and run very close to the RF end of the LPF board.

    All of the bands show about the same output power with a drive of about 28 watts except on 10 meters. At 10 meters the High SWR protection circuit is tripped and shuts down the linear. In this case I think the trimmer cap across the input of the transmission line transformer needs to be adjusted to bring down the internal SWR. So that is the second issue to figure out and fix.

    Outside of those two things everything else is peachy. The linear turns on when the 7300 is powered up and shuts down when the transceiver is powered down. As mentioned before the AUTO band selector works great. Even the ALC line from the transceiver to the linear seams to be doing it's job and keeping the power output to 700 watts or less if I drive the input above 35 watts.

    From here I need to address those two issues, finish the markings for the front panel, make and mount the two walnut side panels and do some performance testing. In that that regard, I did a quick measurement of the 2nd order harmonic at 80 meters and at 500 watts it is down 45.6 dBs below the fundamental frequency. Oh, and I need to put it on the air to see if all of this work was worth it.

    Ripley
     
    AF6LJ, N1OOQ, KG7SWP and 1 other person like this.
  7. N1OOQ

    N1OOQ Ham Member QRZ Page

    Awesome. It's great when all that work finally comes together.
     
    AF6LJ likes this.
  8. KD8UYQ

    KD8UYQ Ham Member QRZ Page

    Thank you Tom,

    Today I took the whole setup to the Radio Museum here in West Virginia where I am volunteer tour guide and handyman for Show and Tell and to get input from the guys there. One thing that was pointed out to me is the placement of the Transmatch board in the circuit. According to the schematic the Transmatch board is connected between the output of the PA board and the input of the Low Pass Filter Diplexer board. That is fine if you want to know the forward and reflected power at that point but doesn't tell you what the antenna is seeing. Made perfect sense so I am going to rewire the Transmatch board so that the power measurement is taken at the output of the LPF board.

    Also, I am going to take the power supply out of the cabinet and run it in isolation to confirm that there is an alternative path to ground that is shunting the current shunt thus the reason the current display is showing me a lower indicated current draw from the power supply.

    I will report back.

    Ripley
     
    KG7SWP likes this.
  9. KD8UYQ

    KD8UYQ Ham Member QRZ Page

    Hello everyone,.

    I fixed the issue with the unstable readings on the LCD power and SWR display by shielding the wires running from the transmatch board to the LCD display. The shield was made from a piece of RG-58 coax cable where I remove the center conductor and dielectric insulator leaving only the copper braid and the black outside jacket. The two wires that send the detected forward and reflected DC signals to the display board were run inside of the copper braid and the braid itself was grounded at the end nearest the transmatch board. That looks like it did the trick. Also, I rewired the input to the transmatch board so the output of the LPF diplexer board is feed to the input of the transmatch detector and the output is connected to the UHF connector where the antenna feed line is attached. Now I am measuring the forward and reflected power at the feed line to the antenna. That is the good news.

    The bad news is removing and isolating the switching power supply from the linear amps metal cabinet made no difference on the current display. I still see no reading of the PA's idle current where I did before when everything was mounted on the mock up board. Also, when the linear is putting out 500 watts the current display tells me that the current draw is 11.5 amps. I don't believe that for a minute. Worse yet, all this was done in the evening after a long and busy day at the Radio Museum. While reinstalling the power supply back into the cabinet I wasn't paying attention to the AC wiring to the power supply and had a predictable outcome when you are working tired. Pushing the power switched was greeted by a bright flash, tripped house circuit breaker, blown fuse for the power supply, and arc welded power switch. A new power switch is on order even though I took the burnt power switch apart and cleaned the contacts so that it works again.

    It gets worse.

    You remembering me stating that when I select the 10 meter band the High SWR protection kicks in and shuts the linear down. I now have found that the high SWR protection circuit activates on the 160 meter band as well. This evening I was in the process of looking into why this was happening when I noticed that the LED for the 15 meter band was intermittent. While probing around with a plastic tuning stick I must have pushed something into something else on the selected band LED board because the next thing that I know the LCD and LED displays are greatly dimmed and I can smell that something is getting really hot. It looks like I killed the +12 volt regulator IC.

    Great!
    Enough for tonight. I will replaced that IC tomorrow and try not to shoot myself in the foot for a while.

    Ripley
     
    AF6LJ likes this.
  10. N1OOQ

    N1OOQ Ham Member QRZ Page

    Ugh. Unavoidable part of experimenting, unfortunately... occasional mishaps. I have little doubt you will recover from it!
     
    AF6LJ likes this.

Share This Page

ad: PortRo-1