Keysight 36100A series PSUs

[bson]

Just got an announcement for the Keysight 36100A PSUs in my mailbox...  Looks pretty good for a $900 street price (checked at testequity) - small, LXI, can measure down to microamp current, and yeah - just really cute little supplies.  If they're as low noise as their other supplies I'd call it a winner (coupled high-z noise around 8MHz being the main weakness of my DP832, plus the Rigol is *huge* for its power rating).

[BloodyCactus]

yes I am glad to see Sense on front of the unit!

[neslekkim]

plus the Rigol is *huge* for its power rating

I bet the Rigol also would be small if it was only 30-40watts..

[D3f1ant]

Big thumbs up for sense, and another for sense mounted in the front. They look cute. Perfect for microcontroller low voltage/power stuff, not sure it needs all that connectivity...until I want to build it into a testbed I guess.

[con-f-use]

They are on the keysight ebay store for 900$ before VAT so no competition for the DP832.

Key Specifications
Output rating (0 to 40 °C)

    Voltage: 20 V
    Current: 2 A
    Power: 40 W

Programming accuracy

    Voltage: 0.05% + 7 mV
    Current: 0.05% + 1 mA

Ripple & noise (20 Hz to 20 MHz)

    Voltage rms: 2 mV
    Voltage peak-to-peak: 30 mV
    Current rms: 1 mA

Readback accuracy

    Voltage: 0.05% + 5 mV
    Current: 0.05% + 1 mA

Full Datasheet

Did I mishear or did he say something about an ultra low current range with single digit micro (not mili!) amp accuracy?

[jitter]

Yep, he said that accurate micro amp measurements are possible. He doesn't say that it is actually displayed on the OLED screen, so perhaps it's only possible to see with software running on your PC.

There's also a teardown video of it (not taking it apart won't do on the EEVblog  , hope Keysight will send one to Dave for a closer look):


He says something about hybrid circuitry, if I understood correctly SCR followed by linear post regulation. Am I to understand that the rectification is somehow microprocessor controlled and that's how they get 40 W max in such a small case without going switch mode?

[mij59]

Yep, he said that accurate micro amp measurements are possible. He doesn't say that it is actually displayed on the OLED screen, so perhaps it's only possible to see with software running on your PC.

There's also a teardown video of it (not taking it apart won't do on the EEVblog  , hope Keysight will send one to Dave for a closer look):


He says something about hybrid circuitry, if I understood correctly SCR followed by linear post regulation. Am I to understand that the rectification is somehow microprocessor controlled and that's how they get 40 W max in such a small case without going switch mode?

A "linear" pre-regulator is not that of an innovation, e.g. its also used in  U8002A  power supply, instead of scr's, mofet's are used.
The pre-regulator is kind of a low frequency (50 or 60 Hz) switch mode regulator.

[jitter]

Does that mean that you regulate the amount of AC that's going to be rectified, depending on the demand?

[mij59]

The output voltage of the pre-regulator is controlled by the changing the phase angle, similar to what is used in a dimmer circuit.

[H.O]

Damn, that is some serious high tech, they are really moving into new territory with that donut shaped thing called a toroidal transformer which has now  become readily available for use in power supplies....

Seriosuly, it looks like a very nice and compact unit but $900+VAT for 35-40W (depending on model) is a little bit on the high side - even considering you get that cool donut shaped thing. 

[jitter]

Damn, that is some serious high tech, they are really moving into new territory with that donut shaped thing called a toroidal transformer which has now  become readily available for use in power supplies....

 

I don't really understand that remark coming from a company like Agil... eh... Keysight. More like they avoided toroids in the past and finally decided to embrace them (a bit like an early Queen album, it would say "No Synthesizers!" somewhere on the cover/in the booklet, how that changed later on...).

Seriosuly, it looks like a very nice and compact unit but $900+VAT for 35-40W (depending on model) is a little bit on the high side - even considering you get that cool donut shaped thing. 

IMHO too expensive for DIY, especially since you can get PSUs that get the job done for far less. Clearly aimed at companies that will very likely use these in (semi-)automated testgear for low power products. I wonder if they're going to introduce dual tracking supplies in this kind of form factor too.

[HighVoltage]

It does look like a nice power supply but the prices seems high.

The comment to the toroidal transformer is funny indeed!

Also, it seems they are catching up with the success of teardown videos.


 

[Rolo]

Very nice PSU, I like the layout of the display. In the clip he is talking about the old day's and SCR pre-regulation. My Delta E030-3 has this type of regulation, still works perfect. It makes a specific sound when under high load, that's the pre-regulator. And no fan needed @ 90W output ! It did get an upgrade on the meter, original it had a analog meter and a switch to select volts or amps.

[Christe4nM]

I have to agree that these supplies are probably aimed at ATE solutions. They seem relatively deep while having a small front size. That is ideal for minimizing rack space.

I compared the ripple and noise specs with the E361x series. The new ones are less good especially comparing peak voltage ripple. That enhances my impression this is not the best for the bench or R&D use.

Price is way too steep in my opinion for a single output supply. For about the same price you have the old(er) e3631a which is three channels, programmable, and maybe even quieter in the ripple and noise area.

The display is too small imo. I like a big text telling me in one quick glance what the voltage and current values are.

I have to say I'm impressed by the small size though. In addition it seems the user interface is intuitive and well thought through. That really stands out to me as so many others get that wrong or at least convoluted and clumsy.
 
I do dislike the rounded edges of the front though. I relate that too cheap Chinese rubbish somehow. Not to say it looks cheap, the rest makes up for that.

Well that's my first impression.

[PioB]

Hi,

does anyone own one of these already and would have some experiences to share? We're looking into getting one for some testing at work; are there any labview drivers available, I didn't find anything on the NI site... (Keysight has MatLab drivers only, it would seem.. http://www.keysight.com/main/techSupport.jspx?cc=US&lc=eng&nid=-35673.1138231&pid=2585379&pageMode=OV )

Thanks for any information

[dom0]

I have to agree that these supplies are probably aimed at ATE solutions. They seem relatively deep while having a small front size. That is ideal for minimizing rack space.

98.5 mm height, 106.4 mm width, 367.7 mm. Pretty much perfect for putting four of these side by side in a 19" rack carrier.

[jitter]

Hi,

does anyone own one of these already and would have some experiences to share? We're looking into getting one for some testing at work; are there any labview drivers available, I didn't find anything on the NI site... (Keysight has MatLab drivers only, it would seem.. http://www.keysight.com/main/techSupport.jspx?cc=US&lc=eng&nid=-35673.1138231&pid=2585379&pageMode=OV )

Thanks for any information

No experience, but my guess would be that you can access the PSU from Labview through the LAN (LXI) interface using the IVI drivers.

Look here, and you'll find that Keysight added support for E361xx series to the existing E36xx series drivers.

This passage mentions NI Labview support (funny, they use both "Agilent" and "Keysight" in this document):

Version: 1.2.1.0      August 2015

Introducing the AgilentE36xx IVI Driver for the Keysight E3600 Series of Power Supplies
---------------------------------------------------------------------------
  This instrument driver provides access to the functionality of the
  AgilentE36xx through a COM server or ANSI C API which also complies
  with the IVI specifications. This driver works in any development environment
  which supports COM or C programming including Microsoft Visual C++, Microsoft
  .NET, Keysight VEE Pro, National Instruments LabView, LabWindows CVI, MATLAB(32
  bit only) and others.


Supported Instruments
---------------------
   E3649A, E3648A, E3647A, E3646A, E3645A, E3644A, E3643A
   E3642A, E3641A, E3640A, E3634A, E3633A, E3632A, E3631A
  E36102A, E36103A, E36104A, E36105A, E36106A

[soldernerd]

I have ordered a E36103A several weeks ago and finally had it in the mail today. I don't usually do reviews but since the thing just came out and few people seem to have experience with it I've shared my first impressions here: https://soldernerd.com/2016/03/11/keysight-e36103a-lab-power-supply-review/

I hope Dave will do a thorough teardown / review sometime soon. Until then, if you have any specific questions just ask and I'll try to answer them.

Cheers
Lukas

[bitseeker]

Welcome to the forums, Lukas. Well done review with nice clear photos.

[bitseeker]

In the clip he is talking about the old day's and SCR pre-regulation. My Delta E030-3 has this type of regulation, still works perfect.

Yep, pre-regulation and ribbon cables with locking connectors. Sometimes the old ways are better. Like the old Power Designs 3650S: 180W, passively cooled, and only 500uV rms ripple. But it's not tiny.

[soldernerd]

Just a brief update. I've now used a E36103 (20V, 2A) and a E36104 (35V, 1A) for about half a year and really started to like them. I used them whenever their output power was sufficient which is most of the time for me.

However, last night I somehow managed to fry my E36103. It was providing something like 1A at 16volts when suddenly the 22V OVP kicked in and the PSU started making very unhealthy sounds. I immediately powered it off but the smell already indicated that some components already went up in smoke. When I tried to start it up again it would first not boot at all and then it started ok but would no longer work.

I contacted Keysight by email and got a very fast response that they would repair the unit. I really have no idea what killed it. So I hope I quickly get it repaired under warranty.

best
lukas

[bitseeker]

Sounds like a pass transistor shorted out letting the unregulated voltage through.

[LaurentR]

Unexpectedly,
Keysight released a new version (E36100B) in the past month.
I didn't look in details, but beside the obvious black front panel, there is some tightening of the ripple specs.



Output ripple and noise (20 Hz to 20 MHz)
36100A
Voltage RMS 350 µV 2 mV 4 mV 5 mV 15 mV
Pk-Pk 10 mV 30 mV 60 mV 100 mV 150 mV
36100B
Voltage RMS 350 µV 0.8 mV 1.2 mV 1.55 mV 2.5 mV
Pk-Pk 8 mV 15 mV 20 mV 30 mV 40 mV

[HighVoltage]

Interesting.
It looks good in black but for some reasons I have not accepted black as the color for Keysight.

[Mr. Scram]

Interesting.
It looks good in black but for some reasons I have not accepted black as the color for Keysight.

This units looks better in black, but I am too still getting used to it.