First impressions

First step was to get a circuit diagram for the amplifier from RF parts. It was not until I studied it that I realised that it was for an MLA2500 and the circuit for a MLA2500B were slightly different. However a posting to the Amplifier reflector got me a copy of the correct circuit diagram. The 8875s were taken out of their sockets and their heaters checked and found to be OK. Lack of continuity between the appropriate electrodes was also confirmed. The six electrolytics were removed and tested. Two were short circuit. It was found that their innards were the source of the red gooey substance, so they would all be discarded. Closer examination of the EHT board was then made and about 50% of the diodes were short circuit, along with a similar number of equalising ceramic capacitors. The time delay relay and the switching relay were checked and found to be fully functional. Next step was to check the transformer was functioning. All windings were disconnected and measured on an AC voltmeter and found to be present. It would appear that a replacement EHT board would be a major step towards recommissioning the amplifier

Initial thoughts were to make an exact copy of the original EHT board (Fig 1) replacing all the components verbatim. However I have never been a great fan of strings of diodes and equalising capacitors, especially as half mine were dubious, so the opportunity was taken to replace them with a pair of K2AW PCB mounted diode blocks, each rated at 6kV 1A. The circuit was initially developed as shown in Fig 2, removing some of the redundancy in the existing board. The PCB was then developed and produced, the track layout and component overlay are shown in Fig 3 and 4. The installed board is shown in Fig 5. Note that only the tracks and contacts used in my amplifier were reproduced; many of the connections on the edge of the board were unused and D16 and C31 were not fitted.

Fig 2. Circuit for replacement for Original EHT board

Fig 3. PCB layout for replacement for Original EHT board

Fig 4. Component overlay for replacement for Original EHT board

Using small pieces of aluminum angle the board was mounted in the original boards mounting holes. The only adjustment needed was to remove a piece of the board to stop it fouling on the power meter terminal (top left corner of PCB in Fig 5). The Input and output SO239 RF connectors were both found to have charred insulation and replaced. The safety RFC across the output was also found to be open circuit and replaced.

Fig 5. Replacement PSU board Installed

A visit to our local electronics emporium in Dallas produced six 560uF 450V working capacitors but no mounting clips. A search in the junkbox produced 6 suitable mounting clips (obtained at the microwave update surplus tour in Silicon Valley in 2001). Each Capacitor had a pair of 100k 2W resistor across it. The capacitors were mounted on a 0.25” piece of Plexiglas (Perspex) using nylon screws and nuts in view of the high voltages around, see Fig 7.

Fig 6. Replacement EHT smoothing Capacitors

The remaining circuitry was then hooked up, still leaving out the tubes. The amp was then switched on and amazingly 1800V was seen on the anode voltage meter. After 90 seconds the EHT voltage tap selection relay actuated and 2400V was seen on the meter. Whilst looking closely for signs of stress the amplifier was monitored for 30 minutes and pronounced OK, before turning off the 110V AC.

The front panel ssb/ cw tune switch was put in the tune position to restrict the EHT to 1800V. With the amplifier input connected to my FT1000MP and the output connected through a power meter into a 50 ohms termination, the tubes were inserted and power re-applied with no sign of tube stress. The PTT terminal was earthed and a tiny amount of standing current seen (presumably class C?). The amplifier was then tuned on 14MHz and found to produce 550W. Similar amounts of power were seen on the other HF bands. Next the switch was put in the ssb position and the output measured. Power increased to 700W on all bands; however the all lamps on the front panel were dimming seriously on key down, suggesting a sagging 110V AC supply.

Having a 220V AC supply wired into the shack for the 432MHz EME amplifier, I changed the taps on the back panel to run the amplifier off 220V. The opportunity was also taken to make a replacement for the missing cover for the terminals to prevent getting electrocuted. Now the amplifier produced around 950W on all bands, I left the shack with a sense of achievement.