Agilent 54845B won't power on - missing +15 BIAS

[DrNefario]

I have an 54845B that won't power on.  It has a Celestica 0950-4191 power supply which is very different fron the one in the 54845A scope.  I am following the Power Supply troubleshooting section of the 54845B service manual.  By using the 220 Ohm resistor to override the Remote Inhibit signal, I can get the unit to basically turn on the power supply, front panel, fans, etc. and the voltages on the power supply measure OK.  However in addition to the inhibit signal, I have to cycle the connector P9 on the Fan board (remove and reinsert) to get to this state. 
I have measured all the resistances for the main power signals to ground and they are OK.
I have determined that the 5 VSB signal is on (LED on PC-MB is lighted) when the AC power is connected and unit is OFF. 
The main thing that seems to be missing is the +15 V Bias signal.  The service manual does not mention where this is generated and it is missing from the resistance check to ground table.
The unit was working fine last week when it bacame unresponsive to the power button.
I would appreciate it if anybody could help with any information regarding the +15 BIAS signal.

[DrNefario]

More troubleshooting info - I was tracing the signals between the Fan Board and the Acquisition Board and measuring voltages.  In the process of doing this I cycled the P9 connector on the Fan Board several times in quick succession as I thought I heard something like a relay clicking.  After doing this maybe 5 or 6 times within less than 5 seconds the scope booted normally and presented the option to continue or run setup.  So I ran the setup for the MB and set all options according to what the service manual said for the Motorola VP22 MB.  Upon save and exit it started the normal scope mode and everything works normally.  I even powered it off with the scope Power Button and was able to turn it back on - as long as the AC power cord was not interupted.  It would not start again if the AC power was interupted and I had to "jiggle" the P9 connector again. 
I made all voltage measurements again when it was running and everything was in spec EXCEPT the +15 V BIAS line on the Acquisition Board which measures 5 VDC (as does the Rem inhibit TP).  So it seems that the main power supply is OK and it appears that there is something wrong with the interlock inhibit path.  It would be good to know what that test point measures in a normally operating scope.

[m k]

Seems that 5VSB is for multiple positions and one has a capacitor that can't keep on from that direction.
Maybe P9 is overloading and disconnecting it will allow that unknown capacitor load and release what it is blocking.

No idea how +15V BIAS should be.