Why do new scopes and other test gear have soft power switches?

[Gary350z]

Why do new scopes and other test gear have soft power switches?
Why would you want you test equipment to have a standby mode?
I am very curious.

Examples:

Keysight  DSOX2012A Oscilloscope


Rigol DM3058E Digital Multimeter

Rigol DG1022 Arbitrary Waveform Generator

Possible reasons:

1. Could it be to reduce the load on the internal battery that powers the real time clock.
I have read that keeping a desktop computer plugged in removes the load on the internal BIOS backup battery, and if the computer is not plugged in, the internal battery will die much quicker.

2. Quicker boot times? I don't think so because there is still a lengthy boot time.

What do you think?

One other thing:
On some scopes the internal real time clock battery is soldered in (Keysight and Tek), and sometimes buried way inside the unit. When the battery dies you have to rip the scope apart and desolder the battery. These batteries are typically CR2032 lithium coin cells.
How long does the battery last?

[CJay]

Keeping reference oscillators warm and stable would seem to be a good reason for a standby mode.

But I doubt that's the reason, I suspect it's mostly to bypass long cold start times for the operating systems that an awful lot of test gear seems to have nowadays.

If Lithium batteries were the reason then you'd make the batteries user replaceable surely?

[Wuerstchenhund]

Why do new scopes and other test gear have soft power switches?
Why would you want you test equipment to have a standby mode?
I am very curious.

Various reasons:

- The instrument has an OCXO frequency reference which needs to be kept at a constant temperature even when the instrument is switched off

- The instrument requires a controlled shutdown (i.e. PC based instruments), which is done via uses soft-power-down

- The instrument needs to be capable of being activated by an external signal, i.e. over LAN

[nctnico]

- The instrument has an OCXO frequency reference which needs to be kept at a constant temperature even when the instrument is switched off

My Keysight 34461A DMM doesn't seem to keep it's reference heated when in standby.

Anyway, soft power switches aren't new in the test equipment arena. A lot of Tektronix gear (TDS500/600/700 series and DAS9200) from the early 90's has soft power switches.

[brabus]

Hard switches are perceived from the customer as old and cheap.

Not to mention that, in absence of a family-felt soft switch, a significant percentage of users would not find out how to turn on the unit. 

[AndyC_772]

That's not really applicable in this case, though. The power switch in a Keysight 2000X or 3000X scope feels just like a proper, old-fashioned mains on-off switch, but it isn't.

[stmdude]

Modern scopes run operating systems. Operating systems have filesystems that needs to be shut down gracefully.

Same thing that happened with computers when Windows 95 was released.  Suddenly, all PCs grew soft power-switches.

[KL27x]

Hard switches cost $$. Why not save money?

[rx8pilot]

Hard switches cost $$. Why not save money?

That is what I think - it's cheaper to implement on top of allowing a managed shut down of the system, closing out files etc.

[JPortici]

Modern scopes run operating systems. Operating systems have filesystems that needs to be shut down gracefully.

Same thing that happened with computers when Windows 95 was released.  Suddenly, all PCs grew soft power-switches.

"Now you can switch off your computer"

[rx8pilot]

Windows 95. 

[Hydrawerk]

Most GW Instek scopes have a real power switch.

[Len]

Windows 95. 

Oh come on, it's not like Windows is the only OS that wants you to shut down gracefully.

[rx8pilot]

True, but I think it was the only one i can remember with the message 'You can now safely power down' since the OS did not have control of the power supply.

Sent from my horrible mobile....

[xrunner]

Why do new scopes and other test gear have soft power switches?

Well, if you asked Rigol, it's so they can make the LED behind it slowly change brightness for that little bit of extra bling.

[Alex Eisenhut]

It's not a new thing:

[tautech]

It's not a new thing:

Nice, I WANT  a 2901 now. 

Has it kept in reasonable Cal Alex?

[Alex Eisenhut]

It's not a new thing:
Nice, I WANT  a 2901 now. 

Has it kept in reasonable Cal Alex?

That's a pic of someone else's unit, but mine is OK. Very disappointing OCXO, it is a simple on-off thermostat and you can see the frequency deviation tracking the heater... On average it still reads close to 10MHz. Good enough for calibrating time bases. My 500MHz output is dead however.

[6thimage]

- The instrument has an OCXO frequency reference which needs to be kept at a constant temperature even when the instrument is switched off

My Keysight 34461A DMM doesn't seem to keep it's reference heated when in standby.

Anyway, soft power switches aren't new in the test equipment arena. A lot of Tektronix gear (TDS500/600/700 series and DAS9200) from the early 90's has soft power switches.

I couldn't resist replying to this as I've spent a fair bit of time learning about the 3446xA's hardware - you're correct, the reference isn't heated when in standby, the power to the measurement board is completely turned off (by a TI regulator that is right next to the mains input). With the LM399 (the voltage reference used in everything except the 34470A), it is detrimental to keep the reference powered and there is very little advantage to keeping it heated (considering how quickly it warms up). To get any advantage, the measurement board would have to be fully powered (i.e. current flowing through the resistors and op-amps powered), and then it is only likely to age the components for allowing a user to get accurate readings straight away.

The power only goes to the front panel, but not to the main processor. There is an 80C51 microcontroller (NXP LPC932), that is used to control the power state and it, along with the LED, are the only thing powered when in standby. When turned on, the microcontroller informs the main processor (the ARM based ST SPEAR320s) that the power button has been pressed, if it doesn't shutdown within a set time (a second or so), the microcontroller cuts the power. If the multimeter ever goes into its screensaver mode (when the screen dims) you can wake it up by pressing the power button, because of it being a soft button. As the LAN interface is part of the SPEAR320s processor, you can't wake the multimeter up remotely, so it is only there for one reason ...

And this is where I believe that Wuerstchenhund is right, it is very easy to damage modern test gear by just pulling the power, where as soft buttons allow for a graceful shutdown. With frequency counters and spectrum analysers their internal oscillators are kept powered - which is why their specifications say they need to be powered for a large number of hours before their frequency measurements are accurate. In oscilloscopes there is no high accuracy oscillator, so there is no reason for them to be kept powered. The same applies for the vast majority of test gear.

To go back to the 3446xA, they can be turned off at the wall and it is very unlikely you will cause any damage to them - but there are advantages to the soft button. On shutdown, the multimeter saves its current state to the STATE0 file, so it can be returned to the same state on power up - if you pull the plug, this file isn't saved (or more precisely, isn't updated). If the multimeter is writing to a file, the soft-off allows it time to finish the write - whilst this might not work for USB files, it at least protects the calibration data and system files.

[mmagin]

It's not a new thing:

Similarly...
But ironically it's always connected to my 10mhz GPSDO.

[karoru]

Modern scopes run operating systems. Operating systems have filesystems that needs to be shut down gracefully.

Same thing that happened with computers when Windows 95 was released.  Suddenly, all PCs grew soft power-switches.

Infinivisions from the Windows 95 era just copied the software to ramdisk on startup sequence to avoid need for graceful shutdown - you pressed the button and scope just went 'pop'

Soft power switches grew on PCs with introduction of ATX standard, but AT cases with clunky switch were still used for good few years after that. And even nowadays most nice (i.e. >20$) ATX power supplies still have additional hard switch at the back. Really, 5-10A rated power switch costs half a buck in one's local hardware store, it isn't hi-tech;) Switch on the back on test equipment would really be nice - I can have all the T&M zoo connected to the mains and don't worry about the standby for less often used equipment.

[Fungus]

Hard switches cost $$. Why not save money?

Because most of them have a hard switch around the back as well as the soft switch.

Modern scopes run operating systems. Operating systems have filesystems that needs to be shut down gracefully.

But ... 'scopes boot from flash and aren't constantly writing to a file system.

Nope. None of the above.

I'm going with: Power-on times.

These scopes 'boot' faster from sleep mode therefore the users are happier.

(and not just software booting - some of them have internal voltage references that need to warm up)

[Mechatrommer]

But ... 'scopes boot from flash and aren't constantly writing to a file system.

how about user's settings that are changed and still stored in RAM? if you have DS1000Z you'll know how good Rigol is at "not saving" the user's setting at disgraceful shut-off.