NEW Keysight HD3

[simone.pignatti]

Also, it only has a 40MHz channel bandwidth selection, not 20MHz. Why?

That's twice the bandwidth, bigger is better.

But not comparable anymore with all the 20 MHz bandwidth PSU noise measurements out there.

It seems that even that 40 MHz is an digital filter. Not an analog filter in actual front end. Actual order of lowpass filter is not specified.

Each analog input channel has a 40 MHz BW Limit option that lets you disable or
enable a bandwidth limit (using a digital filter). Limiting the channel bandwidth
lets you remove unwanted high frequency noise from the waveform and increase
the number of bits of vertical resolution.


Definitely any measurement out there that was specified with 20MHz will not come out the same on this scope.

HD3 has hardware bandwidth limits 5, 10, 20, 50, 100, 200, 350, 500 MHz, 1 GHz (selectable)
At the moment is a general setting for all channels, on a next FW upgrade will be available for each single channel.

[DaneLaw]

Options - Licenses. [#369]

[tszaboo]

Also, it only has a 40MHz channel bandwidth selection, not 20MHz. Why?

That's twice the bandwidth, bigger is better.

But not comparable anymore with all the 20 MHz bandwidth PSU noise measurements out there.

I think the joke went right over your head, even though I put a clarification the next line...

[tggzzz]

...
 (Even if I can't buy it because Keysight refuse sell it to people like me).

Why you want to buy directly from Keysight? 
You can buy from a distributor (which can offer you support in case you need).

Search for "Keysight support" here and elsewhere, e.g. one current thread at https://groups.io/g/HP-Agilent-Keysight-equipment/message/146632

I know about, but you said that they will refuse to sell. If you really want something from KS you can have it through an distributor like  Batterfly for ex.  @simone.pignatti . You can request support to them, they will open a case to KS as a business and get the support. KS cannot refuse them.

Will the specific distributor still be in business when I need them?

In the UK distributors are Mouser, Farnell, RS and a few other companies I haven't heard of. (RS sells Keysight scopes that are £990k+VAT) Do you think those distributors will give you effective support? RS certainly fails on support for their basic services such as calibrating a voltage source - their staff can't even comprehend that it is necessary to state the uncertainty in their measurements!

Keysight appears not to supply businesses unless they also have a website. If they will make that kind of idiotic equivalence, then what will they be like in the future?

[CRTbrain]

Can anyone confirm the HD3 uses a 14-bit ADC, or is it really the same 10-bit ADC used in MXR/UXR locked into 14-bit high res mode?   Historically, Keysight has had more technology info about new ADCs and ICs in new scopes.  I don't see anything like this.   

[Keysight DanielBogdanoff]

yes, confirmed it's a 14-bit

[maxwell3e10]

First noise measurement.
50R 500uV 1ms 1M point 100MS/s 1GHz

I looked briefly at the file, there are a couple of problems. First it has only 62500 points in it, not 1M.  Perhaps some decimation happens in saving. Also 100 MS/s rate is not sufficient to get accurate noise spectrum up to 1 GHz, so it becomes aliased. You can reduce the time/div until the sample rate is maximum or see what happens if one turns on the bandwidth limiting filter.

[nctnico]

Can anyone confirm the HD3 uses a 14-bit ADC, or is it really the same 10-bit ADC used in MXR/UXR locked into 14-bit high res mode?   Historically, Keysight has had more technology info about new ADCs and ICs in new scopes.  I don't see anything like this.    (Attachment Link)

It could make sense Keysight uses the 10 bit ADC from the UXR and oversamples to get more bits. 256Gs/s / 80 = 3.2Gs/s. Oversampling 80 times gives over 3 bits on top of the 10 bit. Make it 14 because many buildings in the US don't have a 13th floor either.

[jusaca]

I think the joke went right over your head, even though I put a clarification the next line...

Ohhh my bad, I did not get that line 

[Zucca]

1GHz hardware shipped on all models.

https://www.eevblog.com/forum/testgear/topic-keysight-hd3-series-licence-have-anyone-tried-to-hack-that-scope/

Please Dave respond with:

Nope, no one has done it yet.

to continue the good tradition.

[2N3055]

Also, it only has a 40MHz channel bandwidth selection, not 20MHz. Why?

That's twice the bandwidth, bigger is better.

But not comparable anymore with all the 20 MHz bandwidth PSU noise measurements out there.

It seems that even that 40 MHz is an digital filter. Not an analog filter in actual front end. Actual order of lowpass filter is not specified.

Each analog input channel has a 40 MHz BW Limit option that lets you disable or
enable a bandwidth limit (using a digital filter). Limiting the channel bandwidth
lets you remove unwanted high frequency noise from the waveform and increase
the number of bits of vertical resolution.


Definitely any measurement out there that was specified with 20MHz will not come out the same on this scope.

HD3 has hardware bandwidth limits 5, 10, 20, 50, 100, 200, 350, 500 MHz, 1 GHz (selectable)
At the moment is a general setting for all channels, on a next FW upgrade will be available for each single channel.

Simone,
I am quoting User manual.

What you are saying is actual Global BW limit that is also a digital filter, specifically a Hires mode where BW limit is specified in a same way R&S does.

No analog BW limit filters of any kind are mentioned anywhere. I presume channels have 1GHz AA filters before ADC but that seems to be it..

If said digital filters per channel (that are coming) are made to mimic classic 20MHz filters by it's cutoff characteristics than it is fine. Apart from fact that signal that enters ADC  is full BW all the time, so on signals where there is significant amount of energy above 20MHz we cannot use filter to make ADCs life easier...

Best,
Siniša

[electr_peter]

Dave can test what exactly the effect is of 40MHz channel filter and 5/10/20/50/100/200/350/500MHz/1GHz global filter. Requires only high BW noise source and FFT. Filter drop-off could be brick-wall or smooth

[Anthocyanina]

ok, so, this weird space next to the analog inputs. i initially thought it would be for the AWG and that the originally leaked photos were of a model without AWG, but since it's been confirmed the AWG is on the back, the space next to channel 4 has to be for something, right? i mean, it could be just a design choice, since not many of the front panel segments are aligned with much else, like the cavity for the channels not being aligned with the display at all or the control knobs, or it not being symmetrical, the lip at the bottom in particular not being equally wide on both sides is weird... anyway, they included MSO in the model number, so that together with the space there is making me wonder if they are planning an MDO model and that space there is for an SA input? 🤔🤔🤔

[tszaboo]

ok, so, this weird space next to the analog inputs. i initially thought it would be for the AWG and that the originally leaked photos were of a model without AWG, but since it's been confirmed the AWG is on the back, the space next to channel 4 has to be for something, right? i mean, it could be just a design choice, since not many of the front panel segments are aligned with much else, like the cavity for the channels not being aligned with the display at all or the control knobs, or it not being symmetrical, the lip at the bottom in particular not being equally wide on both sides is weird... anyway, they included MSO in the model number, so that together with the space there is making me wonder if they are planning an MDO model and that space there is for an SA input? 🤔🤔🤔

Usually that was the space for external trigger.

[Zucca]

or a fifth channel? 

[Anthocyanina]

ok, so, this weird space next to the analog inputs. i initially thought it would be for the AWG and that the originally leaked photos were of a model without AWG, but since it's been confirmed the AWG is on the back, the space next to channel 4 has to be for something, right? i mean, it could be just a design choice, since not many of the front panel segments are aligned with much else, like the cavity for the channels not being aligned with the display at all or the control knobs, or it not being symmetrical, the lip at the bottom in particular not being equally wide on both sides is weird... anyway, they included MSO in the model number, so that together with the space there is making me wonder if they are planning an MDO model and that space there is for an SA input? 🤔🤔🤔

Usually that was the space for external trigger.

but that's also in the back!

[electr_peter]

That space is obviously reserved for credit card reader - SW license activation would be way quicker

[Zucca]

[nctnico]

Does it support Ali-pay, Apple pay and Google pay? Tap your phone to have 10 minutes of protocol decoding.

[Martin72]

To be honest, I don't understand why people are so bothered by the free space.

[tv84]

I don't know if it's the lighting or the color, but the plastic looks cheap. And that space adds to it.

[mikeselectricstuff]

To be honest, I don't understand why people are so bothered by the free space.

It does look like something is missing - very annoying for the wavegen to be on the back & not  here

[KungFuJosh]

I don't know if it's the lighting or the color, but the plastic looks cheap. And that space adds to it.

True, but can you believe how far oscilloscopes have come?? It's only 2024, and already we have a $22,000 scope with a huge 10-inch screen! 🤯🤯🤯

[Martin72]

I have to admit that the look of the whole device is rather unfortunate.
And the fact that the screen is rather small, considering the price range, is even more irritating.
But perhaps there's a deeper meaning behind it that we can't yet grasp.
Oh yes, I just love the multiple measurement windows.
Even the small Rigol DHO800 has them, albeit not perfectly, but still.
I can only dream of something like that.

[fcb]

There's something unfinished about the front panel design.  Like it's been rushed out, or the tooling modified at the last minute or a major rejig of the PCB.

My guess is that it was a 6 or possibly 8 channel design that got it's wings clipped late on in development for some reason (thermal/cost/performance or they needed more physical width per channel) with the missing input channels would have extended all the way to the right.  The button's have then been spaced more to take out the spare real-estate.  The thick black bezel around the display also makes me think they had bigger screen initially.  When I look at the molding design and button layout of an MSOX4104G compared to this HD3 I wonder what the real story was behind the development.  Anyone torn one down yet?