new? Keithley PSU 2230-30-1

[Circuitous]

Speaking of new PSUs, this one from Keithley looks interesting.
http://www.newark.com/keithley/2230-30-1/power-supply-bench-triple-range/dp/28W2312



Anyone have this yet, or know anything about it?
Given the price point, is it really made by Keithley?

[Jad.z]

Is it just me or does anyone else find it very suspiciously similar to the ones made by Tek.

[Smokey]

Keithley = Danaher Corporation
Tek = Danaher Corporation

[rbola35618]

It also looks similar to the BK 9140

RB

[EEVblog]

Smell very much like a rebadge of the same company that does the BK Precision supplies:
 
and also by Atten:
http://www.atten.eu/power-supply/atten-atm8811-programmable-power-supply.html

Maybe an exclusive for Danaher, but very likely not their design or manufacture.

Dave.

[plesa]

I ordered it month ago. I going to post some internal photos and test results.Price was slightly above 1k USD.
It should be delivered at the end of January.

It seems to be  similar  to ITECH power supplies ( front and rear panel as well).
http://www.itechate.com/en/products.jsp?id=53&sortid=002

[free_electron]

Keithley and Tektronix are now in-breds ( as opposed to pure-breds ) as they both belong to Danaher. And that psu  is an OEM from onehunglo they simply slap their name on. probably an Atten or something.

[T4P]

It's a maynuo. Not atten 

[plesa]

I just received the power supply and make some basic measurement.
The voltage and current readout is really fast accurate ( max 2mV/1mA out).
Supply also measure voltage on output but only when it is switched on.
Rise time - 150-180ms (with and without load) one peak, above spec.
Fall time - 15-20ms ( with load) ~1s without load!!
Noise about 30mV pk-pk.

The outputs cannot be switched ON/OFF independently ( only through menu can be output disabled).
The keyboard layout is quite unusual and it takes me some time to remember that zero is on the right side.
The rotary knob does not have the confirmation capability ( value must be confirmed by enter).

Inside unit it is quite overcrowded ( let me know if you want teardown photos).

[free_electron]

TEARDOWN !!!!!

[M. András]

wow that noise on a 1k usd ps?

[robrenz]

wow that noise on a 1k usd ps?

Probably not measured with a differential probe setup. That is probably mainly ambient noise coupled in the measurement. If not, that power supply should be returned.

[plesa]

Here are photos from teardown.
I cannot recomend this power supply, I know that this is not genuine Keithley....but there are better candidates for bench supply than this.
 - The rotary knob wiggle to much, cannot be used for confirming value, keyboard layout is not standard.
 - Inside there are resistors,TO220 close to each other without any isolation, fuses in heat shring tube.
 - The fan regulation in 15% load oscilate and it is quite disturbing in lab ( better will be constant speed probably).
 - VFD constrast is not as good as you can see on Agilents.
 - Keithley also does not provide schematic to their products ( => repair cost when the warranty expires is equal to new unit)
 - Made in China. On their premium product there is only stickers "Made in US" over the "Made in China" sign in cover ;-)
 - Rigol and Agilents are currently cheaper ( Agilent provide 1272 DMM for free)
 + Fully isolated outputs
 + It is smaller than RIGOL or Agilent PSU

[plesa]

Another set of photos.

[T4P]

One will never know when danaher decides to shift production to china to milk more money out of buyers! It's not that it's not "genuine"

[plesa]

As far as I know they moved completely the production to China already. Including their high-end stuff.
On the website it is visible, I probably received one of the first pieces of 2636B which was assembled in US, because there is sticker over the "Made in China".
Also the 6485/6487 are currenly made in China. But all of them are genuine Keithley, the power supply is the first exception.

[codeboy2k]

wow that noise on a 1k usd ps?

Probably not measured with a differential probe setup. That is probably mainly ambient noise coupled in the measurement. If not, that power supply should be returned.

I was thinking the same thing.  I thought it was going to be a switcher, with a linear post-regulator or something like that, judging by that noise.  There's 30Mhz noise with amplitude bursts at 100Khz or so.  Until I saw the photos, and realized it's a full linear PSU.

So, if it's not coupled ambient noise, and it's not probe or scope noise, then it should go back.

@plesa, measure PSU noise differentially, preferably with an active probe that can remove the DC offset at the probe head. This will allow you to set the scope input to DC coupling and use the most sensitive vertical range on your scope(usually the lowest mV/ or uV/division). If you don't have a differential probe, use the 50-ohm mode of your scope and about 1V output or 1.5Volts from the PSU (1-1.5V @ 50 ohms is 20-30ma current). You can use a female BNC to banana jack at the PSU, and a male-to-male 50-ohm BNC cable.  At 1V or 1.5V DC coupled input you will likely be able to use the scopes offset to bring the signal into the center of the graticule. This will again give you the maximum sensitivity from your scope and allow you to use the full dynamic range available from the scope's ADC.

The problem with AC coupling is that it can often remove LF components, such as slow deviations or LF ripple.

In the picture (PARD) =  Periodic and Random Deviations




Also, I noticed you are measuring the PSU noise with bandwidth limiting on?  I wouldn't normally do this, as I want the full BW, since the noise and it's harmonics is all over the place. But PSU noise is often spec'd with a BW limit..so I understand why you are doing it, just make sure you are not limiting it too much, and once your signal path is clean, you can be sure that you're not seeing any coupled HF noise that's not really there, and you can loosen up your BW limiting to try to capture more of the real PSU noise. (This makes more sense on switchers running at Mhz frequencies)

Cheers!

[robrenz]

Just to add to codeboy2k's comments,  make your scope connection to the PSU and have it plugged in but off, not output off, totally off. What you see on the scope now is all the coupled ambient noise that your setup connection and unpowered PSU are picking up.  Make adjustments in setup to reduce the noise levels in this mode first. Try to get a triggered trace  not just rolling noise.  Save a trace of this as your baseline noise.  Now turn on the PSU and you are now seeing baseline noise plus the PSU noise. If your triggering is still working you can do math to subtract the stored baseline noise from the current signal. (assuming your scope is digital and has these functions)

Here is an example with differential amplifier on an analog scope.

[codeboy2k]

Here is an example with differential amplifier on an analog scope.

Good point on baseline with the power off and cables connected at the supply 

As for that link, nice setup ! I'm still looking at eBay every few days to find a PD supply at a decent price.

[dom0]

What a mess inside!

[luizrenault]

Hi. Does anybody knows where I can find the service manual of this PSU?