Heathkit IB-1100 nixie frequency counter restoration

[MrBungle]

Another thing in progress I thought I'd share. Bought this from ebay recently as non working because I sold one and thoroughly regret doing it as I miss the glow of the nixie tubes!

As supplied, all digits were lit. Opened it up and found leaked capacitors and a very badly bodged in massive crystal from 1955 and ceramic trimmer instead of the expected smaller crystal and piston trimmer. Replaced the capacitors with suitable modern equivalents and powered it up. Still all digits lit. Started with the frequency reference and checked to see if it was oscillating. It wasn't. Pulled out the first IC and applied the TTL output of my function generator to the input of the divider network and I got all zeroes. Attached a signal source to the input BNC and no counting. Fortunately that was just the wire had come off the BNC connector. Quick solder job and confirmed the rest works fine which is a miracle! I think the person who did this bodge had used a 100W iron to do the soldering as the tracks had disintegrated entirely. Due to the virtual non availability of both a suitable crystal and damage to the board, I decided to build a replacement 1MHz reference oscillator from scratch. This is a 74HC00 driving a 4MHz crystal and divided by two 74HC74 steps. This can be tested independently then and qualified before installation.

New board temporarily mounted for testing



Victory



More soon when I get some time...

[MrBungle]

I won.

I installed the replacement timebase I made and it worked out of the box straight away, apart from the KHz range which it was latching the display through making it unusable. A quick look at the schematic pointed to an oversight. The 7400 I had removed that was the timebase originally was also used elsewhere in the KHz range timing generation. I snipped off the pins for the gates that were used for the oscillator which has been removed (being TTL these aren't high impedance so no issue with oscillation) and plugged it in to the socket, then routed the new timebase into the socket and to ground. This cured it nicely.

New timebase installed:



Hooked up to a simple xtal oscillator as my signal generator is dead...

MHz range:



KHz range:



Next clean up, replace the bypassed fuse (which was protected by a secondary 13A fuse - ugh) and calibration.

[MrBungle]

Just in case anyone else comes across this. The fuse specified is a 1/4A 3AG fuse. These are available in the UK from RS/eBay under a 6/6.3 by 32mm 250mA fuse with T characteristics.

[Chaparal]

MrBungle,

maybe you provide us with schematic of the mods and timing modul ?

Peterr

[MrBungle]

I didn't actually do any schematics. I went straight to PCB. Basically it's a standard CMOS 74HC00 based crystal oscillator with a 4MHZ crystal divided down to 1MHz. I wouldn't do it this way to be honest - I only did because I had the parts around. A single 74hc4060 and 8MHz crystal is a better option. The only mod was removal of all the old crystal oscillator section (with aid of the service manual) and inserting this in the start of the frequency divider chain. The original 7400 used for the crystal oscillator had one gate used for some timing so I snipped all legs off other than for that gate and the Vcc/Vee leads and put it back in the socket. The new timebase is plugged into the IC socket and soldered to the ground plane. This was done via crude signal tracing with a DMM