Just checked my '12A and I see the same 10.4W standby, 12.5W or so active but no channels enabled. Hasn't been turned on all day so definitely isn't to finish cooling down after a heavy workout or anything.
Pretty weird. I think I can just about hear the fan (something of a pain to look at the back with my setup) but it's extremely quiet, I wouldn't have noticed if I wasn't looking for it.
I have one of these on order. This kind of "standby" consumption is really not acceptable in this day and age...
Hopefully, it's just a firmware change. It's not like the supply has an OCXO to keep warm.
Maybe they did it to reduce boot time? But I agree that for a bench supply, that unlike a system power supply is usually only used for eight or so hours per day, this is ridiculous.
I have one of these on order. This kind of "standby" consumption is really not acceptable in this day and age...
~92kWh a year, for me about EUR 20,- extra costs. What is about 8% of my total consumption per year for my whole house.
...
I'll talk to that team and see what they say about the standby power consumption. If it's an issue, you can always run a hard switch with a power strip.
Daniel, just checking in - was there any update on the team with regards to the power consumption issue discussed earlier in the thread? TIA!
Looks like a nice basic bench supply.
A few nits that I would pick:
1) In the video they talk about current measurement down to 100uA, but the display is showing a resolution of 1uA, and with no load connected, the current is non-zero. I hate it when a result is displayed with more resolution than accuracy - which happens with 93.85326% of test equipment vendors.
2) Someone mentioned not liking the 'kindergarten colors' for the outputs - I do like this. In my job (validation engineer for a semiconductor company) I will often have several different test boards for the same part, and I color code the cables for power to make it easy to figure out which supply goes where - by matching cable colors to the channel numbers it will make life easier.
3) I like the supply sequencing feature, but I suspect that it is not possible to tie multiple supplies together and sequence across two or more power supplies - I often have to sequence more than 3 supplies.
Looks like a nice basic bench supply.
A few nits that I would pick:
1) In the video they talk about current measurement down to 100uA, but the display is showing a resolution of 1uA, and with no load connected, the current is non-zero. I hate it when a result is displayed with more resolution than accuracy - which happens with 93.85326% of test equipment vendors.
The unit I had on loan was accurate to less than 10uA so it makes sense to show a resolution of 1uA.
Resolution beyond accuracy can be useful for observing relative changes. Is the current drifting up or down? Accuracy beyond resolution is wasted.
The only problem is people too lazy to check the equipment specs before taking a critical measurements.
My E36312A has arrived
, eight weeks from order they must be popular?
Keysight have a new promotion that includes this series of supplies (promotion code: 6.023), however as mine was ordered in August I wanted to take advantage of the previous promotion, code: 6.015. This is valid up to 31st Dec 2017 for items purchased between 1st April and 30th September 2017.
I think there is a problem with the webpage. Choosing either E36311A, E36312A or E36313A from the "Model Number" drop-down list results in the all important "Serial Number", "Purchase Date" and "Purchased From" textboxes disappearing.
Would a kind soul try the page and let me know of similar action?
www.keysight.com/find/BIG5BenchClaimYes, I'd like claim the U1242C
to replace my "one hung low" special.
It seems all good things come to those who wait
Now they throw in a free U1282A dmm. I just pulled the trigger on a E36313A. That is the 160W version but when it comes to accuracy the specs are very close to the 80W version. For me the 80W version isn't going to cut it for a number of projects I find on my bench regulary.
Edit: these PSUs seems to be popular. I have to wait nearly two more months before it will be delivered!
Anyone with one of these units notice an audible hum when the unit is "off"? Mine (E36312A) has a low hum when I'm sitting in a quiet room with everything off; it's not noticeable at all when other gear is on though. It seems to alternate periodically from a steady hum to alternating on/off every couple seconds or so.
It's definitely still "on" even powered down. That's a soft power switch. I think it draws about 10W on standby too.
I have the E36312A and there is no hum at all.
But there are a few firmware issues I have come across that are very annoying.
Sometimes the PSU just stops responding and I need to unplug the instrument for a power cycle, since the soft power switch is then also not responding.
What I find also annoying:
The voltage turning knob can increase the output by 1mV per indent.
But the real output is not increasing by the same amount for each indent.
In other words, there is a big difference between the set point the actual readout and what is measured externally with a different meter.
Hmm, seems like a firmware bug as well.
This could be much more precise.
I think the hum might be the fan occasionally spinning up?
This video (which nihtila posted on the contest thread) highlights a bunch of bugs and options for improvement, perhaps including your issue HighVoltage:
Why would you want to use the knob to adjust the voltage? On PSUs with a keypad I just punch the number in. Much quicker and easier.
I think the hum might be the fan occasionally spinning up?
It's don't think it's the fan, I can hear the fan ramp up when I power it on and the hum remains in the background. It's not a huge deal but when I'm on my nearby PC (which is dead silent) with nothing else in the room on, there's just a slight, annoying hum. I was just curious if it was typical of these power supplies. If everything is turned off I might start flipping off the power bar because I don't see any reason for it to be humming and drawing 10W all the time.
Why would you want to use the knob to adjust the voltage? On PSUs with a keypad I just punch the number in. Much quicker and easier.
Usually I also just enter the numbers on the keypad and it works really well.
Finally a Keysight PSU with a keypad !
But the other day I needed to ramp up the voltage in mV steps and used the knob, expecting every step to be 1 mV change. But then to my surprise this was not a linear increase. May be it is nitpicking?
The voltage turning knob can increase the output by 1mV per indent.
But the real output is not increasing by the same amount for each indent.
In other words, there is a big difference between the set point the actual readout and what is measured externally with a different meter.
I've been eyeing these supplies for a while so I'm curious, how big a step size are you actually seeing with your 1mV setting changes?
The front panel controls are all spec'd with 1mV steps, but the remote programming resolution for the E36312A is 0.36V for the 6V output and 1.5mV for the 25V outputs. Perhaps these represent the real hardware resolution (14 bits???), and all you are seeing is the quantization error of mapping the 1mV steps onto this.
Of course if you are seeing larger steps then this wouldn't be an explanation at all...
A while ago I ran a little test with the E36312A ch1 hooked up to my 34465A, and recorded the difference between measured voltage on the '65, and both the programmed and readback voltage on the 36312A. I did it over a few spans but this first one is the only one with mV steps. Also attached is over the whole 0-6V span. It's not bad all around, error tends to be within a fraction of a mV.
The voltage turning knob can increase the output by 1mV per indent.
But the real output is not increasing by the same amount for each indent.
In other words, there is a big difference between the set point the actual readout and what is measured externally with a different meter.
I've been eyeing these supplies for a while so I'm curious, how big a step size are you actually seeing with your 1mV setting changes?
The front panel controls are all spec'd with 1mV steps, but the remote programming resolution for the E36312A is 0.36V for the 6V output and 1.5mV for the 25V outputs. Perhaps these represent the real hardware resolution (14 bits???), and all you are seeing is the quantization error of mapping the 1mV steps onto this.
You have a very valid point here and something very similar is going on with HP's older telecom DC power supplies like the 66311B. The programming and readback accuracy are different. Add the inaccuracies of the circuitry to that and it is clear that changing the setting (=programming) by 1mV isn't going to match the readback reading.
Here's a more direct comparison. Increment in 1mV steps around 3V, shows the PSU readback vs the DMM readback. I think it's pretty close to incrementing by a mV each time you crank the knob by 1mV.
Here is an example of what mine sounds like. Audio quality isn't great but you can hear it. The only other thing running is my obnoxiously loud clock.
I've been using an E36312A casually for the last couple weeks and can confirm that the fan does turn on even if the unit is turned off - but the fan is really, really quiet. So, there's not an audible noise issue for me at all.
I'll talk to that team and see what they say about the standby power consumption. If it's an issue, you can always run a hard switch with a power strip.
Overall, though, I've been loving the supply.
Daniel, do you have any information on potential firmware updates for the PSU?
I'd also like to know if anything will be done to the accelerating V and A knobs.
I have tried the E36313A (considering getting one) and I am not convinced at all this is right for a PSU.
I like having a knob to step the PSU (i.e. with a fixed step). That's how most other similar knobs work (including from Keysight, e.g. on the 33522B AWG - pick a significant digit and step it up or down).
But the whole thing about quickly adjusting the voltage, I don't get it. I use the keyboard for that. It's actually faster than accelerating to the right area and decelerating at the right time and most likely going back some.
I have an E36313A and get a similar audio buzzing noise even when powered down (powering up can still yield the buzzing noise). I tried uploading the audio file but it's too large. Suffice to say that this happens even when the unit is powered down (but plugged in). My symptom is it lasts for 1-2 seconds, then goes quiet. It doesn't sound like the fan turning on (why should it anyway, when it's been off for a long lime). It sounds like some 60 or 120 Hz buzzing that is very audible. I was able to capture the current and voltage going into the supply and the current spikes nearly 1A when the buzzing starts, then it drops back down. This is with the supply turned off. Now that I see someone else with this problem, I believe my situation is not unique and that there may be environmental situations that could be interacting with the power supply circuitry (such as plugged into a power strip with a surge protector, etc). Who knows, but it's very strange for sure. I have also come a cross many UI bugs, most of which I believe have already been discovered and reported on this forum, including total instrument lockup, which needs to be fixed fast.
adamgreig - thank you for your comprehensive and excellent review. Do you know whether Keysight engineering has seen your video yet? They need to and I encouraged the Keysight support team to have a look as well. I have run into many of the same issues (although you've found many more). I've reported them to Keysight as well. The egregious bugs like locking up indicates very poor software QA and there is no excuse. If something is in a rack or bench top, it's not easy to unplug it always to reset it. I'm not clear how these could have passed the kind of rigorous process I used to know (I worked for the original HP T&M division for 13 years). Makes me wonder if this is an OEM product and Keysight relied on third-parties to design and test it. I hope not for the price. They really need to improve the software to make the UI more usable, less prone to errors, and obviously fix the bugs impacting usability. Also, I remember now if you mentioned it, but when you save and recall settings, it does not recall the Logging setup either. Maybe this isn't a bug per-se, but it's a pain because I want to be able to save and recall the instrument setup, including the logging feature. It can take a bit of time to reprogram the logger setup, especially when the unit completely hangs up!
In addition, I have experienced the same issue as you with inadvertently adjusting the current or voltage not realizing that the active area was something else. As a result, I accidentally applied 20V to a diode (good bye diode) when I was trying to current limit to 20 mA. This is simply due to the way the UI works. Yes, my fault, but too easy to make an error.
I also have the issue that others have reported of a periodic buzzing noise of the unit (even when powered down). During this condition, it draws an extra amp of current for a couple of seconds, then goes back to its quiescent state. It sounds like a fairly loud "buzz" or oscillation of the power supply, even while turned off. Keysight wants my unit back for evaluation but after visiting this forum I see that at least one or two others have reported similar issues with an audible buzz or noise. In my case, the fan is not on -- as it happens even overnight with nothing turned on (I've captured the transient in a power data logger, showing the instrument side of the load).
Anyway, great review. You captured so many of my frustrations with this unit. To be clear, I'm an avid supporter of Keysight/Agilent/HP T&M equipment. But with this unit, I feel like they lost their magic touch and that with some software improvements, they will have an outstanding piece of gear. Otherwise, it's a frustrating experience and I'm tired of pulling out the cord to reset it. No excuse, Keysight! Put your best on this to fix these things and please watch this video from adamgreig. Customers don't take time to do this because they want to waste their day. They do it because they love the brand and want the products to work in a way that's intuitive and reliable.