Ricahrd, welcome to Beoworld !

The tuning voltage zener diode sits in a holder bolted to the rear facing side of the tuner
frontend metal casing.
It has thermal coupling in the form of heatsink paste (compound) to keep its
temperature, and with it its voltage, stabile.
Clean off the remains of the hardened and cracked paste and add fresh.

Martin

Thanks Martin,  I wil let you know how it goes when I check that Zener and its heatsink paste.   think I will monitor its voltage over a period as well.

Richard

Martin:

When first turned on, my BM1900's tuner #5 is supposed to be tuned to 90.7 MHz.  In reality, though, it pulls in something like 93.5 MHz.  Over the course of about 5 minutes time, it will eventually slide back down to almost 90.7 MHz.  A little fine-tuning later, and it's back to pulling in the 90.7 MHz very well.

Is this the same problem with the same solution?

Rich

Something like that but I seem to remember that the BM1900 circuit is slightly different.
Check the tuning voltage and you will know more.

Martin

Martin,

Sorry for not getting back to you sooner, and many thanks for the advice.  I eventually got around to pulling the top off and having a good look at the Tuner power supply circuit. 

It is fed from the 60v supply and TR10 (BC212L-B PNP) appears to be the controlling device.  The 22v Zener diode (Item 138) that you described (which is marked ITT ZF22) goes from the TR10 Collector to Ground (0v), while another Zener diode (9.1v  Item 137) goes from +60v to the TR10 Base.

The Base is linked to 0v by a 33kΩ resistor (Item 139) and the Emitter is linked to +60v by a 2.2 kΩ resistor (item 136).   The nominal 22v supply current is therefore also regulated by the action of the 9.1v Zener which biases TR10’s base.   The current supply (at nominal 22v) passes through another 2.2 kΩ resistor (item 143) to reach the top of the set of  0.1 MΩ trimpots that provide the tuning action.  

I removed the clamp holding the 22v Zener diode and also unsoldered the diode from the terminals, in order to carefully clean the diode, clamp and RF casing of any of the old heat-transfer compound and re-make the soldered joints.  You were right, what was left of the compound was dry and cracked and probably wasn’t transferring much heat at all. 

I then “tweaked” the clamp so that it would hold the diode firmly against the casing (it was a slightly loose before) and then re-assembled the diode, clamp & screw  to the casing with new heat-transfer compound .

I thought a long test was in order so I put a couple of meters on to monitor the voltages at the 22v Zener and at the top of the trimpots (lower end of 2.2 kΩ resistor item 143) over time.  The results are below, note that the voltage at the top of the trimpots was varying (oscillating) between the two values, rather than exactly steady.  I assume that small variation is something to do with the switching within TR10, do you have a view on that?.

Switch on (T0)      21.64v                   18.77-18.79v

T0 +18 mins         21.65v                   18.78-18.80v

T0 +80 mins         21.67v                   18.79-18.81v

T0 +3 hrs              21.67v                   18.78-18.81v

T0 +5 hrs              21.69v                   18.79-18.82v

Since then, the FM radio seems to be able to hold tune on a given station for a significant time, whereas before this it used to wander off after 10 minutes or so.  If you know of any possible failure modes in the RF section (electrolytic capacitors etc??) I would be most interested for preventive maintenance.

So it appears that your suggestion about the diode was correct.  Many thanks for that and a happy New Year for 2012.

Regards

Richard