Dave wrote:
>> While I agree with you here, I have to admit
that, if my PDP-11/70
>> had switching power supplies, I'd probably run it a lot more often.
>> I'd *never* make it an irreversible modification, though.
I wrote:
> Unless you've already replaced them with
something else, your
> PDP-11/70 *DOES* have switching power supplies!
Dave wrote:
I'm looking at the schematic for the H744, and I
see an LM723 with a
big fat 2N5302 (Ic=30A) wrapped around it. Now, I'm aware that the
LM723 can be used in a switching regulator topology, but at first
glance that's not what it looks like to me.
I'm somewhat embarrassed to admit that, because I've repaired
probably a dozen H744/H7441 regulators, including one just a few weeks
ago in my newly-acquired PDP-11/50.
Yes, it's a switcher. The switching transistor has a 30A rating because
the rated output of the supply is 25A. The theory of operation is
covered in section 4.4.6.2 of the maintenance manual, EK-11070-MM-002
(May 1979), page 4-59 et seq., and a sample switching waveform is given
in figure 4-39 on page 4-61.
For the KL10, I'd like to upgrade to switchers but I don't want to
replace the entire ECL power supply as is done in the Compuserve mode.
I'd rather keep the front end, which consists of the H760A raw supply
(basically just a three-phase transformer and rectifier), and the
capacitor box (13 ea. 0.3F 15V electrolytics), and replace just the
individual -5.2V 35A and -2.0V 35A regulator modules in the H761
regulator assembly, since those modules can be easily removed and
reinstalled. I've got a design for replacement regulators using the
LTC3731, and it looks good in simulation, but I haven't yet built any.
I wasn't quite sure how to use the LTC3731 as a negative buck regulator,
but one of the apps engineers at Linear Technology showed me how to do it.
Eric