CMC/Marconi BH-30 Owners?

Discussion in 'Amateur Radio Amplifiers' started by VA3DXV, Jun 29, 2018.

ad: L-HROutlet
ad: l-rl
ad: L-MFJ
ad: FBNews-1
ad: Subscribe
ad: Left-3
ad: Left-2
  1. VE9GFI

    VE9GFI Ham Member QRZ Page

    Thursday June 18th

    Now that I have been successful in getting a rather large amount of RF output, I have directed my time to cleaning up the amplifier. The coil L901 is fairly corroded/tarnished. The TUNE and LOAD air capacitors initially had signs of corrosion. This was cleaned up using a paint brush (one that I regularly use for dusting out circuit boards). Just by gently brushing the plates, I was able to get some of the white corrosion off the aluminum, but the bulk of my time was spent on the coil.

    The coil was removed from the chassis. In order to do this, I opened the side panel so that I could photograph all the taps and their positions. I then numbered each tap (on their sleeves) so that I could easily associate each tap in the photo and their positions. When I was able to remove the last tap, the frame that contains the coil was removed and place on my bench. I took the frame apart, and swung open one of the lengths of bakelite, which allowed for the coil itself to be removed. I used Peek Creme Polish and a soft non-abrasive cotton cloth as it seems to do the trick. I followed that with a clean cotton cloth and the coil is now shiny, although there is still some signs of tarnish. I have placed the coil back into its frame, and after work today, I'll continue with cleaning between the loops with a tooth brush and cotton cloth, to see if I can remove the remaining tarnish. Even if I can't make a lot more headway, the current before and after pics do show signs of improvement.

    Before I place the coil and its frame into the chassis, I will slip off the insulators from each tap and clean that metal as well. The band switch S901 is similarly tarnished and I'll need to decide if I clean it and how. I see that only four bolts hold it down, but I'm not sure how this is attached to the motor B901's shaft. But as I see this now, not cleaning S901 would be like not changing the timing belt on an engine, when the engine is already torn apart.

    Greg

    20200617_221003.jpg 20200617_221117.jpg 20200618_082015.jpg 20200618_083519.jpg 20200618_083551.jpg
     
    W1TRY likes this.
  2. VE9GFI

    VE9GFI Ham Member QRZ Page

    Wednesday June 24th

    With work-from-home in a basement devoid of sunlight, and scorching hot days and evenings, development has slowed. But not to a stop.

    I have ordered replacements for all six pilot lamps. I tried to purchase dual LED lamps, but those that fit my budget were 40 mm long and these will not fit where the air capacitor motors are located. Those that did fit were either backordered, or in the range of $20+. So, I purchased the single colour replacements.

    And did re-install the L901 coil last night. I took about 25 photos of the tap locations prior to its removal and I needed those photos and the video to be sure that I reinstalled it properly. If you are wondering, putting it back is harder. But I'm liking the shine, and I'll be re-aassembling the air capacitor section once the lamps are installed.
     

    Attached Files:

  3. VA3DXV

    VA3DXV Ham Member QRZ Page

    Hi Greg, glad to see you got the beast home! Man that coil looks fantastic. Great job. Mine wasn't corroded as yours so I didn't do much detailing, but now that I see the results maybe I should. The silver plating on the rest of mine is pretty tarnished in spots. I'm really looking forward to seeing that assembly when you get it put back together.

    For the band switch, I had to remove the input assembly from the bottom because the airwound coil was damaged, I haven't tackled the tank circuit yet. I think if you wanted to take the upper(tank coil switch) and lower(input switch) off all as one unit you can do so. But from underneath at least, it's a servo driving a rotary switch for the input, connected to the tank coil switch via a semi-square brass rod. To take the input switch off the assembly, there's a roll pin holding 2 shafts together at a coupling. Can't remember if the coupling was on the servo motor itself, or if the servo was able to be moved out of the way first. I can't remember how I managed to get mine off from underneath but I do remember inventing some new words in the process. There was likely an easier way. Anyways when I was putting it back together I tossed the roll pin and used a small cotter pin instead. Much easier for the future. I think since you're re-doing the tank circuit anyways, it might be possible to remove it all as one assembly, and maybe easier.

    If you do separate the two assemblies, you have to pay close attention to the switch positions when you put them back together.

    I think RFParts has replacements(pulls) for R203 and R204

    It's interesting to see yours is missing the same pilot light cap as mine was. I was lucky and found exact replacement light sockets in a box full of electronic junk someone gave me, and I stole the cap off one :) I don't know if those are available anywhere? I did find bulbs for the pilots in the RF deck are available at Allied, part #104581

    I saw your video and I wish I could get my channel switch operating like that. There's either a problem with the actuator/servo on mine, or I really messed up somewhere on my controller. Something was getting hot and it smelled like something scary was about to happen. I removed it and went back to manually operating for now.

    Looking forward to future posts, Cheers!
     
    Last edited: Jul 4, 2020
    VE9GFI and KD2ACO like this.
  4. VA3DXV

    VA3DXV Ham Member QRZ Page

    Here you go,
    Click for full size

    [​IMG]
    [​IMG]
     
  5. VE9GFI

    VE9GFI Ham Member QRZ Page

    @VA3DXV Brian,

    I did replace all pilot lamps in the top section with 12VDC parts from VCC. https://www.digikey.ca/en/products/detail/visual-communications-company-vcc/1092QC3-12V/5965082 . I also ordered the same type of VCC pilot lamps (120VAC versions) for the Filament and HV.

    The air capacitor is not fully assembled yet, so I'll do that first to be certain that all the connections are in place before applying any power.

    Thanks for the photo of the motor control board. I have the rats nest circuit in KiCad and I can position the parts according to your photos. On that part... I wrote a Perl script that reads the radio frequency from HRD and calculates the band position. What I'm thinking of doing, in the long term, is to use a Raspberry PI or Arduino to read the frequency detail from HRD. This in turn would assert the correct ground on the P101 connector, thereby taking care of the band selection. The last step would be to use a DAC to output the proper position voltage reference to the OpAmp in the Motor Control Board that corresponds to the particular position of the TUNE and LOAD capacitors for that frequency. This would replace the potentiometers that exist today in the original Motor Control Board, which essentially provides tuning capability for 10 channels. If I setup an ADC to read the wiper voltage, I should be able to store the proper tuning position during programming via a push button switch. So tuning procedure would go like this:

    - unmesh the TUNE and LOAD
    - set the Motor Control Board to program mode.
    - apply RF
    - dip the TUNE by turning the TUNE knob
    - press the program button to store the voltage reference into memory
    - max RF by turning the LOAD knob
    - press the program button to store the Load voltage reference into memory
    ... and so on until you reach your desired input and output RF levels.

    At this point, your two Motor Control boards will have stored the correct voltage (i.e. position) of each the TUNE and LOAD capacitors for that particular frequency (and power level and plate voltage). When you change frequency, the DAC would then output the proper voltage that corresponds to the position of the capacitors.

    Note that I also mention Power Level and plate voltage. The impedance of the tubes vary with power level and plate voltage. The above description would work if I always wanted to operate at the same power level but to be more accurate, reading those two other values would render a better tuning circuit.

    So, for the present moment, I will be reassembling the air capacitors, now that the pilot lights are in place and I'll run it up to be certain that it still runs as it did a few weeks ago. And then I'll layout the original Motor Control circuits and replicate that part.

    Greg
     
  6. VA3DXV

    VA3DXV Ham Member QRZ Page

    Great, hope you're able to get those boards made up. Can't wait to see what you come up with. I think I'm going to start looking at something to completely replace my load/tune boards and the alc/control board with a custom circuit. I want to eliminate S301 altogether, since my actuator (or whatever actuates it) doesn't seem to work anyways. I don't think I have the patience to troubleshoot it.

    I haven't tried it yet, but basically I'd just leave my amp on channel 1 forever, and then change frequency range assignments at the jumper board instead of changing channels. Unless I really want to keep the antenna switch functionality, I really only care about switching S901 anyways, S301 never needs to move... and I don't think there's anything (maybe?) that would prohibit me from just controlling S901(B901) at the jumper board. I'd have to wire up a rotary switch to channel 1 on the board and try it. I can see the possibility of being able to replace everything in that compartment except the relay board with an arduino/pi contraption. Could even make replacement boards that just slide right in to the factory slots, probably minimal modifications to the rest of the amp.

    Then if I ever figure out S301 and the actuator, well that's no longer a channel selector. It's an independently controlled antenna switch.

    I found the pi makes a ton of RF hash and noise. There's a lot less noisy stuff going on with an arduino, but the pi sure would be nice if it can be silenced.
     
  7. VE9GFI

    VE9GFI Ham Member QRZ Page

    OK - I see what you mean by the S301. Yes, if you had this on channel 1, you would never need to move it again. In fact, I might want to do the same, since my plan is to replace the TUNE and LOAD boards with boards that read the frequency and put out the correct voltage reference to move the Motors. With that setup, the concept of channel goes away.

    I am a long ways from the PI, or Arduino, or Teensy 3.1 controller (I have two in my desk), or _insert good idea here _ .... I'll keep posting.

    Greg
     
  8. VE9GFI

    VE9GFI Ham Member QRZ Page

    Saturday July 11

    We are usually an hour away at a campground each weekend, but we stayed home to celebrate an outdoor birthday party that eventually was cancelled due to thunder showers. And the showers did come (and they are still here.) No complaints, as this is the first rainfall I've seen in at least 3 weeks. We've had mist and fog, but no rain until today, so let it pour.

    This provided me the chance to reassemble the S901 switch, Tune and Load capacitors, and replace the two AC lamps on the PS part of the chassis.

    After a couple hours, I now have the unit back into it's fully assembled and shiny state, with 5 new pilot lamps. My next step is to use my MFJ-259 to measure the input impedance. I have been told that the input impedance is this amplifier's downside. On that point, can someone tell me what is happening at the RF input as it goes through S901B and S901C? It seems to be a shunt L and shunt C, with both values changing with the bandswitch.

    It feels good to have this fully assembled as I was starting to worry that I wouldn't get it back together properly. The one thing the manual is missing is an assembly diagram. The good thing about having taken this apart is that I can point out each layer of S901 and all the components, which will come in handy should I choose to extend the bandwidth to 160 m, and up to 10 meters. I have no current plans for that yet.

    Could someone comment on the filament voltage? The voltage under load (Tubes in), but not transmitting is 5.45 volts, which is outside of the specification of the 3-500Z, but in line with the BH-30 Manual. I read that tubes will last for a long time if you keep them between 4.75 to 5.25 volts, and that if the filament voltage is low, it will poison the tube and if it's too high it's equally bad. What is your filament voltage and should I be concerned?

    The motor control board is now the focus, and I just printed out the photos so that I can use this for my KiCad layout.

    Greg
     
  9. VA3DXV

    VA3DXV Ham Member QRZ Page

    If I recall correctly, mine was about 5.1 v or so. Maybe you have high line voltage there? Using 130v taps? My mains here is 121v most of the time so I wired it for 130v until I get 240 at its location. I imagine you'd have to expect shortened filament life at 5.45 but how much I can't say.

    Did you only take the top end of S901 off? or did you get the top/bottom as a complete assembly?
     
  10. VE9GFI

    VE9GFI Ham Member QRZ Page

    Hi Brian,

    I removed the entire S901 using the bottom 4 screws. I lifted it high enough so that I could flip the tune/load capacitors on its end. While changing the pilot lamps, I pulled on a wire on the motor of the Load Capacitor motor, and the only way I could get access to it was to make room to flip it upwards. Not fun.

    The two transformers are tapped for 115 volts and similarly, I will eventually tap this amplifier for 240V. In fact, I already have the receptacle and wire, but when I opened up the fuse panel, I see that I don't have any free positions to add any fuses (which I understand is a no-no as it means that if one of my fuse positions fails, I can't move a circuit over, let alone add a circuit.)

    My son is an electrician, so I'll have him add a sub panel. Even while just running a dehumidifier, PC, and turning on the HV on the amplifier (not transmitting, just flicking the HV on), I blew the breaker.

    Greg
     

Share This Page