IFR 1000A - HF PLL - Why does it fail?

[warnergt]

The HF phase lock loop on IFR 1000's seem to have a high failure rate.
This is the PLL that moves in 1 MHz steps and lights the left frequency lock light on the front panel.
You will often see these IFRs on ebay with that particular light out.
Mine is out now and I'm guessing that this is a somewhat common problem.
Before I start debugging this, I thought perhaps somebody already knows why these fail and could save me some time.
Why do these HF PLL's fail?

Mine isn't a total failure. If I spin around the frequencies, I can get it to lock on some frequencies (like around 300 MHz).
It seems that something has got weak with time.

[VE3TUH]

Try Kurt Graber from KG electronics. He fixes these older boxes. Could be anything from a power supply going south to a failure in one of the modules.

[warnergt]

Just to follow up, I resolved this problem.

The HF (1 MHz steps) VCO was intermittent. It was a subtle problem and, being intermittent, took a while to find.

The VCO tuner board runs on +11V and -35V. It outputs about 0 to -20V to the VCO which, in turn, ranges from 1200 to 2199 MHz.

My problem was on the power supply and regulator board. It uses two transistors in a current mirror configuration as part of the -35V regulator. One of those transistors was bad. Because of that, it was outputting an unregulated -39V — something that might appear nominal when measured with a volt meter.

So, the tuner was operating with an unregulated input voltage. Sometimes it worked; sometimes it didn’t. Now, it is fixed.