• EEVblog #164 – Agilent Fly To The Moon

    Dave is a tad upset to find out how much standby power consumption the new Agilent Infiniviision oscilloscope draws.

    Be Sociable, Share!

      About EEVblog

      Check Also

      EEVblog #821 – Mailbag

      EEVblog #821 – Mailbag

      More Mailbag! Extended Teardown Video HERE Forum HERE SPOILERS: Military Mystery Item Teardown – UPDATE: ...

      • Cesar

        I would like to see the response from Agilent, they better have a good reason. How can this equipment deals with power consumption legislations?

      • Steve

        Hey Dave I agree that it is kind of strange to put in all that hardware and not put a hard power switch in the unit. The only reason I can think of is that the unit can be woke up remotely via the Ethernet or other type of communication port. You might look in the option menu to see if you can turn off any standby options. That might get the current consumption down to less than a 1 watt. Very common to have this in consumer products but usually the default is off rather than on.

        As a side note, I’m curious why you don’t use something like a Kill-A-Watt P4400 power meter (www.p3international.com). Very cheap, full feature, and they have a plug on the front instead of hardwiring. Seen them as low as $15 each. But maybe no one make a 240VAC model.

        • John

          Probably the for the very same reason he doesn’t have a $3 multimeter – they’re both ridiculously unaccurate.

          • Steve

            Actually it is fairly accurate. At least accurate enough for the type of measurements being made here. Just did a load test comparison a P4400 and my HP 3468A. P4400 read 0.53A and the HP read 0.53805A. Sure if you want 5 digit accuracy, use something else. But if you just trying to check load power consumption, its a great little tool.

        • http://www.eevblog.com EEVblog

          For many years you have not been able to buy a low cost Kill-A-Watt type meter in Australia. Only recently have the come in, and most of them are shit at measuring <10W.
          I have a Silicon Chip meter kit too. Same result. The Gossen meter was just nicer to show on video.

      • http://kj6ead.hackhut.com KJ6EAD

        If keeping the switcher warm to increase MTBF were an issue, I’d think a simple soft start circuit would suffice. “Energy Scar Certified”

      • Coder

        I use Tektronix TDS3000 series scopes. They also have a mechanical switch that you would expect to isolate the instrument from the mains. Yet they also draw about 5 Watts in standby mode and feel warm to touch. I emailed Tek about it years ago – never got a response.

        • inse

          The secondary switch on the TDS3000 is a compromise.
          The scope can be operated by a battery pack which is being recharged even in “OFF” state.
          To be able to use both power sources, you would otherwise need to have two “switches” – a standby button and a power switch.
          (like on PC’s)
          I made myself a cable with a lamp switch to disconnect it from the mains.

      • http://chemicaloliver.net chemicaloliver

        This looks similar to how a standard pc works, on an atx power supply one of the pins is momentarily bridged to ground and the thing breaks into life. This generally uses a few watts when turned off but allows things like wake on lan to work it’s a bit of a waste for the majority of people.

        • John W

          Not as much of a waste as you’d think, unless you like replacing those coin cells that keep the BIOS settings and clock going. +5VSB is a handy feature of modern power supplies, they’re not terribly efficient because I think they’re usually linear regulators (like a 7805), but they’re not terrible.

          Shame the power factor is pretty poor on that pricey scope, though. You’d figure they could spare a little bit of room for a switching transistor/cap/diode for PFC, save a bit on the power bill in the end.

          • http://www.eevblog.com EEVblog

            There is no standby power going to the main board, so it would not help the standby battery

      • Gene

        Looks like its main board does not turn off but rather goes into power-down mode to reduce start-up time. What are startup times of the scope (both cases, turned off by a switch and turned off by disconnecting mains) ?

        Another assumption is that they are having problems with physical turn-off, maybe filesystem gets corrupted or something like that, so they decided to put it into power-down instead so it can properly shut-down in the process. This whole thing looks like last-minute engineered to me. The board has a lot of space to accomodate large mechanical switch, yet they use only small logic-level one. If logic-level shutdown was designed in from the beginning, it would definitely go elsewhere, on the opposite end of the board, on the another board, etc.

      • Dennis

        Well, modern file systems use journaling techniques so that you won’t corrupt it by pulling the plug in the wrong moment, never. So since Agilent doesn’t use Windows 95 on this device, …

        But reducing the system startup time by putting the scope into standby sounds plausible.
        Or it is really some MTBF issue, which would be really shameful for Agilent.

        • Gene

          > Well, modern file systems use journaling techniques

          Only if designers have decided to use it. A lot of Windows CE based platforms just use FAT over flash translation layer. I’m not sure if there is journaled file system for WinCE at all.

      • comox

        Nice one again! I am obsessed about standby power consumption and have a couple of Yokogawa WT210 power meters which I use to analyze appliances and such, which are very accurate down to the 5ma range. I have a Tektronix MSO4032 which has a big mechanical pushbutton switch on the front which I ASSUMED isolated the power supply circuit from the mains when off. After viewing your blog I hooked the MSO4032 up to the WT210 and here are my observations:
        MSO4032 on: 238V (UK Mains), .361A, 77.54W (Mode RMS)
        MSO4032 standby: 241.92V (UK Mains), 3.57ma, 0.0007W (Mode RMS)
        Sorry Dave, but I am not going to take my scope apart and look at the PSU circuitry!

        • inse

          …so what’s the point, comox?
          3.57mA and 700uW don’t match, so the current seems to be reactive.
          I ASSUME it’s the capacitors in the line filter which is located before the switch…

      • Edisonobilly

        I think the speculation about the main board remaining in standby mode still needs to be proven. From what Dave said, the switch seems to cut the output from the SMPS, which I doubt leaves any other voltage supply for the board to remain in standby mode.
        Dave, could you investigate further to know if this is correct? If the switch simply cuts off power from the SMPS, then that would have been a really poor design decision.

        • FreeThinker

          I think Dave confirms in the latest amphour that the switch only kills the output from the psu, so yes a pretty piss poor design

        • http://www.eevblog.com EEVblog

          I have mentioned this a few times now. Yes, the switch kills the output of the PSU. There is no voltage going to the main board.

      • chuck

        What’s the model number of the main’s power meter?

        Can you explain how the power factor thing works? i.e. what 86% @6W vs 3% @20W??

        • Andy

          Do you want Dave to do a video on the power factor or do you want a short answer ?

          Power Factor is the solution of

          cos(phi) = PowerFactor

          So with the PowerFactor and a Pocket calculator your result is an angle, this Factor is important in AC applications (but also HF) as in an AC circuit current and voltage can go out of phase to each other, and this is what that factor describes.

          They can go out of phase because many wound power lines with magnetic fields crossing each other are a coil, even if you call it an electrical motor.

          The background can be understood via Wikipedia:

      • HammerFET

        Madness! I say modify that switch to bring in the mains, then use a micro or a 555 to make a 1 – 2 sec delay to activate the start up.

        It may be worth seeing if the function of that switch is firmware configurable. I’m sure Agilent did that to allow for some kind of LAN wakeup feature or something like that, I wouldn’t be surprised to see a simple option that allowed for the main controller to turn off such peripherals.

        I’m sure Agilent will explain themselves anyway.

      • sionyn

        what version of ce is it using 6.0 use exFAT which is not a journaling file system.

        my guess is the wake up lan too

        remind me of freeviewbox box i came across that uses the same amount of on and in standby.

      • Vlad

        This is hilarious… You should put a lamp switch on it! Hahaha..

      • Daniel

        That’s crazy. This joke would cost me about $20 US dollars a year at $0,36/kwh here.

      • Matthews

        If you pause at 5:08, A picture is worth a million words;



        – Matt

        • HammerFET

          Lol, just figures, and here we are coming up with funky excuses for LAN wakeup 😛

      • Mike in Alaska

        Maybe its for some sort of remote operation capability with the LAN connector? Being awakened to take measurements?

      • Jay K

        Good find Dave. In CA that scope would cost about $7.50 a year just to have it off. …The scope you never stop paying for. Inexcusable.

      • Luciano

        The power supply CCH125 has a remote on/off control
        (a low signal turns the power supply off) and has
        a single 12V DC output. If in the “Off” state there is
        no 12V voltage on the power supply output, then it is
        clear that the oscilloscope is completely off and
        doesn’t have a Wake-up on LAN function.

        CCH125 power supply datasheet:

        See the page 96 of the 2000/3000 X-Series Oscilloscopes
        Service Guide “To check the power supply DC output”.


        * * *

        I’m quite sure that Agilent has evaluated many brands of
        power supply during the development of the oscilloscope.

      • Luciano

        (Sorry it’s me again from Italy).

        Measurement of Standby Power & its Implications

        * * *

        Are we sure that the used Gossen Metrawatt multimeter is capable
        of an accurate measurement of the no load power consumption of
        an SMPS? See page 79 and 80 of its manual.

        What about repeating the measurement with something like that
        or something more accurate?

        Or simpler, ask the manufacturer of the power supply for the
        no load power consumption of this SMPS? (Power consumption
        when the SMPS is turned off via its remote control feature).

      • Steve

        I don’t know about your particular dso, Dave, but I’ve been told before (at shows, no less) that people prefer chunky power buttons and thats why they use them (and from your first video review, that seems true for you too). They don’t however like the ability to cut off the mains power supply feed at any given moment. This gives them the opportunity to do things like save settings when you switch the dso off, rather than everytime you make a change.

        It would be interesting to see if the infinivision is the same. Quick check: Change some of the settings and then exit and pull out the power cord. See if the settings remain after you plug it back in :)

      • Luciano


        (I saw your message posted on April 23rd, 2011 at 18:10, thank you).

        I still think that the values you get are not accurate because
        the multimeter is unable to measure the power consumption of this
        SMPS with the output turned off via its own remote control pin.
        From the datasheet of the SMPS and from the model number
        visible on your video (time 5:06), I understand that this power
        supply has active power factor correction which improves the
        system efficiency. (Regulations in Europe require PFC for SMPS
        with more than 75W). This is a modern power supply so why have
        an active power factor correction when the power consumption
        with the output turned off via its own remote control is so bad?

        It would be good if you could perform the following test:
        Connect a 100W incandescent light bulb to the Gossen Metrawatt
        multimeter and measure its power consumption and write it down
        on paper. (The measurement of this resistive load will be
        very precise, and the multimeter can certainly handle that).
        Now, in addition to the load of the light bulb, add the load of
        the oscilloscope powered off and measure the power consumption
        of the two loads together and see if what you get matches what
        you get when you add together on paper the individual power
        measurements of the light bulb and the oscilloscope. If you get
        a discrepancy that is significant enough, then it is most likely
        that Gossen Metrawatt multimeter is not suitable to measure the
        power consumption of this SMPS while its output is turned off
        via its own remote control pin.

      • Everaldo Luiz Grothe

        Hi, my contribution to the topic.
        I performed a measurement in Agilent DSO-X 2012A that I have with my Fluke 43B.
        I hope I helped.

      • Everaldo Luiz Grothe

        Hi, my contribution to the topic.
        I performed a measurement in Agilent DSO-X 2012A that I have with my Fluke 43B.
        I hope I helped.
        Please note at the beginning I inform Power x 10 so the power is 5.2 W in Stand by mode.