tom,
I'd not use that supply on a valuable rig until fixed. It almost sounds like
just a bad connection, but might cause spikes and damage something.
I suspect something in the wiring to the transistors.

Sounds repairable, if you've got an experienced friend.
Tom Fowle WA6IVG

On Fri, Nov 27, 2015 at 07:46:32PM -0500, Tom Behler wrote:
> Butch, Pat, and all:
> 
> Well, after last night's test of my RS35 transistors, where I discovered
> that the two transistors on the righthand side of the supply became much
> hotter than the left while running the 590 with 100 watts into a dummy load,
> I then decided to take some more power readings.  I took the readings both
> with the internal meter on the 590, and with my TW1.
> 
> You may recall that, the other day, the readings were 75 watts out according
> to the TS590 meter, and 58 watts out according to the TW1, and the supply
> meter read 13.8 volts and said I was drawing around 11 amps.  This time, the
> readings were better, meaning that   My output power readings were much
> closer to full power according to both the TS590 meter and my TW1.
> 
> I then asked my sighted XYL to take a look at the meters on the RS35 power
> supply itself.
> 
> Even though the voltage seemed to be at a constant 13.8 volts while
> transmitting into a dummy load at 100 watts, my XYL noticed that after a few
> minutes, the current readings on the RS35 ammeter began to fluctuate.  They
> would fluctuate anywhere from 14 amps, to 18 amps.  When I checked the
> transistor heat sinks on the back of the supply for temperature, I again
> found the same variation--the left-hand ones were nice and cool, while the
> right-hand ones were almost too hot to touch.  Furthermore, my XYL noticed
> that when I touched the heat sinks, the amperage values on the meter
> fluctuated as well.
> 
> I don't mean to be obsessive about this, but this supply is making me
> nervous.
> 
> I think that I'm just going to bite the bullet and replace it, and then take
> my time working with someone in my local club to see if we can fix the
> supply so I can at least have it as a backup.
> 
> I really would like to continue being able to run both my TS590 and TMV71A
> with my rigrunner, so will be sure to purchase a supply that would be
> adequate for the job.  I know some have recommended another RS35, and I
> might just go with that, since I would prefer a linear supply.  However, I'm
> always open to other recommendations if anyone has them.
> 
> In the meantime, I've got one or two  backup options, including the TS53, to
> at least keep me on the air, both on HF and VHF/UHF.
> 
> Tom Behler: KB8TYJ
> 
> 
> -----Original Message-----
> From: For blind ham radio operators [mailto:[log in to unmask]]
> On Behalf Of Butch Bussen
> Sent: Friday, November 27, 2015 6:09 AM
> To: [log in to unmask]
> Subject: Re: RS35 Transistors
> 
> The hot ones are likely good as they are in series with the load, that is
> why those supplies are called series regulaters.  I'd replace all four from
> the same vendor so you have sort of a matched set.  Should be able to get
> those from mouser or some one.  You will also need some silicon grease.
> When you pull the transisters, watch for the thin mica insolater that is
> between the chassie and transister.  Some times these stay stuck to the
> chassie, or heat sink, and some times they come with the transister.  You
> don't want the transister actually touching the heat sink.  A small bit of
> silicon grease is used between insolator and transister and between
> tinsolator and heat sink to help with heat transfer.  I'll have to see if I
> can find the number, but if I were doing it, I'd go with darlingtons, I
> think e c g 247 or e c g 249 which are the same, but higher current.
> 73
> Butch
> WA0VJR
> Node 3148
> Wallace, ks.