Seeking PSU advice: Rohde & Schwarz NGE103B COM3 or Keysight E36313A.

[jjoonathan]

It doesn't have a display off / screensaver by default and I don't see an obvious configuration for it. I prefer always-on, so I haven't really gone digging. It has brightness control, the ability to lock front panel input, and a mechanical power switch + 4 second boot time in case any of those address your requirement.

[RBBVNL9]

Thanks for the various submitted ripple and noise measurements!!!

Looking at these efforts, at Dave’s video EEVblog #594, and some of my own measurements, I feel amazed how large the effect is of external noise sources (dimmers, radio stations, and more) when using an improper measurement setup.

After all, we are dealing here with voltage sources with very low output impedances, low resistance cables (hopefully), and very low impedance loads (in my own experiments I use a 10 Ohms resistor). My feelings would tell me that this should eliminate most if not all of the effects of such external sources. But (my) feelings do not seem to be a good guide here...

[Fixpoint]

But (my) feelings do not seem to be a good guide here...

Well, except when you are an electronics god, feelings AND measurements are both tricky things. They both betray you. Doing measurements right is so difficult that I am very cautious when it comes to that. I know that I am no expert and that I am likely to make a thousand mistakes. So, that's why I usually don't try to measure high-performance equipment, because for that I would need to be high-performance myself

[justanothername]

A recent discussion with measurements is here:
https://www.eevblog.com/forum/testgear/rs-nge100-voltage-jumping-around/

[Wolfgang]

But (my) feelings do not seem to be a good guide here...

Well, except when you are an electronics god, feelings AND measurements are both tricky things. They both betray you. Doing measurements right is so difficult that I am very cautious when it comes to that. I know that I am no expert and that I am likely to make a thousand mistakes. So, that's why I usually don't try to measure high-performance equipment, because for that I would need to be high-performance myself

Always the same story.
If you want
- very low noise (350uV RMS, 1-2mVpp)
- high stability
you get
- linear mode
- low efficiency

If you want
- high power
- high efficiency
you get
- high ripple and noise
- switching mode

For sensitive analog stuff, like bridges, low noise amps, VCOs, sensor circuit,, detectors, ...
- get a linear to save you a lot of grief

For bread and butter analog or digital
- get a switcher

My choice:
- Rigol DP832 (just look at ripple and noise) for normal stuff (400$)
- Keysight B2962A with filters for real sensitive stuff (AttN: 10K$)

NGXX and Keysight "cheap models" are 1-2K and still noisy.
Dont see why they are attractive from the specs. Pay 2-3 times the price
for just the name with worse specs than second row ? Why ?

[RBBVNL9]

Always the same story.
If you want
- very low noise (350uV RMS, 1-2mVpp)
- high stability
you get
- linear mode
- low efficiency

If you want
- high power
- high efficiency
you get
- high ripple and noise
- switching mode

Fair point. In another threat I yesterday made a similar point: "The NGE is a power supply, designed to significant current over a wide range of voltages. To do so in a compact housing R&S chose to use a switching regulator (offering efficiency and compactness) followed by a linear control circuitry (for stabilisation). Having switching elements inside (like many modern designs do) inevitably will result in some noise on the outputs."

For sensitive analogn stuff, like bridges, low noise amps, VCOs, sensor circuit, low noise amps, detectors, ...
[...]
My choice:
[...]
- Keysight B2962A with filters for real sensitive stuff (AttN: 10K$)

That sounds like a very nice instrument. 

I don't need such a very clean source everyday, so sometimes when I need very low ripple but when precise voltage setting etc. are not important I just take a battery (which is not to suggest at all that it is equivalent to a high end source, but it can serve as a relatively clean source for at least some purposes, and for comparison, at least comparisons to 'noisy' PSUs).

[Wolfgang]

Always the same story.
If you want
- very low noise (350uV RMS, 1-2mVpp)
- high stability
you get
- linear mode
- low efficiency

If you want
- high power
- high efficiency
you get
- high ripple and noise
- switching mode

Fair point. In another threat I yesterday made a similar point: "The NGE is a power supply, designed to significant current over a wide range of voltages. To do so in a compact housing R&S chose to use a switching regulator (offering efficiency and compactness) followed by a linear control circuitry (for stabilisation). Having switching elements inside (like many modern designs do) inevitably will result in some noise on the outputs."

For sensitive analogn stuff, like bridges, low noise amps, VCOs, sensor circuit, low noise amps, detectors, ...
[...]
My choice:
[...]
- Keysight B2962A with filters for real sensitive stuff (AttN: 10K$)

That sounds like a very nice instrument. 

I don't need such a very clean source everyday, so sometimes when I need very low ripple but when precise voltage setting etc. are not important I just take a battery (which is not to suggest at all that it is equivalent to a high end source, but it can serve as a relatively clean source for at least some purposes, and for comparison, at least comparisons to 'noisy' PSUs).

I agree an SMU is an overkill in a lot of cases. Ironically enough, the Keysight B2962A box uses a switcher as a preregulator and a linear postregulator just like many other PSUs (R&S, cheap Keysight). The big difference is the effort that went into cleaning up the signal. AFAIK, there was a teardown of a Keysight SMU at EEVBlog. Another difference is the range 210V/3/30W and the 4 quadrant operation.

It is *much* easier to make a linear PSU with ultra low noise specs than doing it the SMU way. It wastes energy, OK, but for normal lab use this is tolerable if its below 200-300W.

[bson]

If you want to free up bench space, then there's also the Keithley 2230-30-1.  Or if 30mV regulation is good enough, there's an Itech model (forget the model and I'm too lazy to go look).  The 2230 also permits parallel output use, though it might have similar limitations to the R&S.  (The display shows the combined output.)  The drawback of the 2230 is the use of USB, which is too unreliable and impractical IMO for bench use; they could have added ethernet and LXI instead.  As it is, the GPIB option is the best bet which kind of prices these out of competition for me, but for someone who doesn't rely on remote control they might work well.