[ronbryan]

I've managed to fit the MMBF4416 in place of the original round FET. It's the black rectangle at the pointed end of the pcb. The drain tab is extended with a bit of stripped 30AWG wire-wrap wire.

After fitting the FET, the gate resistance to ground was found to be only 2M, instead of 10M. It seemed that the 4.7pF gate-source capacitor C120 was leaking, but when I removed it, I saw there was a thin black line beneath it which was conductive. Fortunately, a quick scrape with a scalpel removed the leakage path.

I experimented with the value of the source resistor for Q100, but quickly found that although reducing it to 1k did reduce the source voltage to the specified 0.05V, this caused too much current to flow through the 1/8W 160 ohm collector load for Q120 and Q130. I ended up with the original 2k source resistor and a source voltage of 0.85V to gnd, which raised the emitter voltage of Q130 to 2.29V to gnd. Even then, the 160 ohm resistor R170 is run at its maximum rating, so not ideal.

I had a further look at the P6201 parts list and found that later serial numbers of the probe are fitted with 0.4W SFR16 Philips/Vishay metal film resistors with the same 3.6mm body size instead of the original BB carbon types.

A quick test of the pcb with a scope and a 10MHz sinewave input showed that the AC amplifier circuit was working, but instead of being near unity gain it was down by 50%. The output co-ax matching resistor, a tiny wire-ended 50 ohm 'pellet' resistor R190 tacked across 2.2pF peaking capacitor C160 had fallen to bits, leaving just a 1k trimming resistor R185 in parallel as the cable matching component. This 1k, when loaded by my 10x scope probe impedance (at 10MHz), was causing the signal loss.

Without a replacement 'pellet' resistor or anything like it small enough, I compromised and removed the 2.2pF SM speed-up capacitor and soldered a 1/8W 51 ohm C3 metal oxide resistor in the space created.

That finally got the AC amplifier working and I replaced the pcb in the probe body and confirmed it worked when connected to the DC-LF amplifier and DC-HF amplifier in the box at the scope end.

There is some adjustment to be done at the 'box' end to correct a 90mV trace offset when the probe is switched to DC coupling, but this is covered in the manual, so hopefully will be straightforward. I don't have the fast rise pulse generator, sampling scope, precision termiator etc for proper calibration, so I will see how I get on with it as it is.

I've attached a photo of the tiny wire-ended 'pellet' resistor (with the green dot) before it fell apart and would be interested if anyone has used these 50A10L parts, apparently made by CTS Corp in the USA.

Ron