NEW Keysight HD3

[shabaz]

..I think the biggest change has been that a LOT more scope users are taking advantage of FFT functionality and that's good for everybody

I wonder if the result of that will be that the new FFT capabilities on 'scopes might even drive SA sales. I'm already curious about the R&S SA's, having seen the MXO 4 capabilities. I currently use a Keysight SA, and an R&S FPC1500, which is a great instrument of course, but is more entry-level.

MXO 4 (and presumably HD3/other 12-bit 'scopes if their FFTs are as good) can also do neat stuff like combining two LISNs, so that you can quickly add or subtract and view the spectrums separately for common-mode and differential-mode noise, and if it's responsive then you'll see a lot as you manually tweak things, way before the product is even at pre-compliance level of readiness. And very easy to use remote programmability with a few lines of Python and view the data and overlay limit lines or run pass/fail tests frequently (pre-Pre-Compliance?) while developing a product.

(Screenshot taken before the log-frequency-scale feature came in).

[tautech]

You see, up until now I've treated my (8-bit MSO5000) oscilloscope, at the advice of others, as a "look but don't measure" type of device. Being able to do more with a scope was something I didn't think was possible. They were just not designed for making accurate measurements, or so I was told.

12+ bits changes all this with ±0.5 % DC gain accuracy now the norm.
Add to this DSO's that provide detailed Stats and quite precise measurement of both axis is now possible.

[David Hess]

Linearity and other errors reduce the ENOB (effective number of bits) limiting accuracy.  At low frequencies where the ENOB is higher, low frequency noise in the input buffer dominates.  There are some other things going on as well like setting time; most instruments will never settle to their digitizer resolution, and high resolution instruments will not even come close.

Yes and what's that got to do with the price of encoder knobs?

CRTbrain claims the horizontal decimation will affect vertical resolution, but provides no evidence that is so. Do you have some examples or evidence of horizontal decimation changing the vertical resolution or accuracy?

Horizontal decimation has a statistical effect on vertical resolution by increasing noise, but it is insignificant compared to the sources of error that I listed.

[hpw]

..I think the biggest change has been that a LOT more scope users are taking advantage of FFT functionality and that's good for everybody

I wonder if the result of that will be that the new FFT capabilities on 'scopes might even drive SA sales. I'm already curious about the R&S SA's, having seen the MXO 4 capabilities. I currently use a Keysight SA, and an R&S FPC1500, which is a great instrument of course, but is more entry-level.

MXO 4 (and presumably HD3/other 12-bit 'scopes if their FFTs are as good) can also do neat stuff like combining two LISNs, so that you can quickly add or subtract and view the spectrums separately for common-mode and differential-mode noise, and if it's responsive then you'll see a lot as you manually tweak things, way before the product is even at pre-compliance level of readiness. And very easy to use remote programmability with a few lines of Python and view the data and overlay limit lines or run pass/fail tests frequently (pre-Pre-Compliance?) while developing a product.

(Screenshot taken before the log-frequency-scale feature came in).

May one day, as using 4 channels, some PN could be done by SW.
Even before cross correlation using two channels and rtHz scale too 

But IMHO, as HD3 with 14 bit shows SFDR as 79dB (not that bad as Keysight SA) may a better SA as Siglents (in the 45dB range).

May the other DSO companies will come up hopefully soon with an 14 BIT ADC too.

[2N3055]

Horizontal decimation by skipping samples simply creates exactly the same effects as sampling with that slower speed, so undersampling.
Some measurements will be impacted more, some less..

[nctnico]

I'm under the same impression, and shops where I was getting quote for an SA immediately wanted to sell me a EMC precompliance test bundle.

Probably because it's often a really cost-effective combination   Seriously.  The cost of failing one full compliance test is usually significantly more than the cost of a spec an and a precompliance bundle.  We have lots and lots of very happy EMC precompliance bundle customers.

IMHO the money for an EMC test suite on a spectrum analyser is better spend on tools like a LISN, low-Z probe, RF sniffer probes and a TEM cell. The reality is that unless you spend hundreds of thousands on an EMC test chamber and calibrated antenna's (basically building an EMC compliance lab), your measurements will never be accurate. So whatever an EMC suite brings, it won't be of much help.

Using 10kHz and 100kHz RBW you'll be close enough to 9kHz and 120kHz and with averaging you can see whether a peak smears out or not. So I don't really care about having quasy peak measurements or the exact RBW on a spectrum analyser.

Among other tools I do have a TEM cell which is a huge improvement over trying to measure emissions with an antenna in an office. Recently one of my designs had to pass EMC testing somewhere in Asia and the locals added some other items to the test setup which caused a fail for some tests. Based on my own testing I could tell (not suggest) they had an issue in the test setup which turned out to be the case. A TEM cell is also very useful (together with RF generator and amplifier) to do immunity testing. Being able to test up to 50V/m with and without modulation can tell you a lot about the stability of a design (at high power levels, do use a shielded TEM cell though).

And then there is the lo-Z probe. Using an RF sniffer probe can be helpfull but I tend to use a lo-Z probe + DC block to just probe signals in a circuit and look at it on the spectrum analyser. In the end a sniffer probe can lead you to a general area, but not the individual signal.

IOW: spend money on the tools.  Most, if not all, of the pre-compliance tests can be done using a modern >8bit DSO with FFT and decent spectrum analysis controls.

[tycoon_9]

In my opinion, a new Keysight HD3 is not worth buying. There are too many spec downgrades, such as Waveform intensity grading, Sampling rate, etc. It is ridiculous to see the waveform intensity grading reduced from 256-level in MSO7104B -->  64-level in MSOX4104A --> 16-level in HD3.  I think Keysight gonna make a 4-level waveform grading intensity in a newer one. Geez, this is suck. The sampling rate is 3.2 Gsa/s? , downgraded from 4 Gsa/s. As of today, it should even be 5Gsa/s in this price range.  For those wanting to buy an HD3 at $10,000, getting a used EXR258A at $11,000 is better. Anyhow, EXR258A has a super long booting/start-up time.   I love previous Keysight products, but new products are not up to the standard versus the money spent.  https://www.keysight.com/used/us/en/oscilloscopes/exr258a-u08-2466971

[mikeselectricstuff]

t is ridiculous to see the waveform intensity grading reduced from 256-level in MSO7104B -->  64-level in MSOX4104A --> 16-level in HD3. 

16 level was a misprint in the datasheet - it's 256 - mentioned above

[electr_peter]

The sampling rate is 3.2 Gsa/s? , downgraded from 4 Gsa/s.

3.2GSa/s for all 4 channels in HD3 vs 5GSa/s 2ch / 2.5GSa/s 4ch in X3000T/G. With new fancy 1GHz passive probes (optional) full BW can be used on 4 channels at once.

[Martin72]

The scopes are now listed by dealers, including Meilhaus.
I got the prices including VAT for private customers from them:

HD302MSO-200     9258€
HD302MSO-350   15375€
HD302MSO-500   19602€
HD302MSO-01G  23717€

HD304MSO-200    10220€
HD304MSO-350    16336€
HD304MSO-500    20563€
HD304MSO-01G   24678€

The prices for the basic models with 20Mpts memory, even serial decoders have to be bought separately... Ridiculous.
Speaking of the options:

HDO300-050 Memory 1668€
HDO300-100 Memory 3337€
HD3WAVEGEN 1335€
HD3SECURE 1113€
HD3EMBA 2781€
HD3AUTA 2781€

Bandwith upgrades:
HD3BW-001   2-channel HD302MSO from 200 MHz to 350 MHz 6117€
HD3BW-002   2-channel HD302MSO from 200 MHz to 500 MHz 10344€
HD3BW-003   2-channel HD302MSO from 200 MHz to 1 GHz     14459€
HD3BW-004   2-channel HD302MSO from 350 MHz to 500 MHz  4226€
HD3BW-005   2-channel HD302MSO from 350 MHz to 1 GHz      8342€
HD3BW-006   2-channel HD302MSO from 500 MHz to 1 GHz     4115€
HD3BW-007   4-channel HD304MSO from 200 MHz to 350 MHz  6117€
HD3BW-008   4-channel HD304MSO from 200 MHz to 500 MHz  10344€
HD3BW-009   4-channel HD304MSO from 200 MHz to 1 GHz       14459€
HD3BW-010   4-channel HD304MSO from 350 MHz to 500 MHz   4226€
HD3BW-011   4-channel HD304MSO from 350 MHz to 1 GHz      8342€
HD3BW-012   4-channel HD304MSO from 500 MHz to 1 GHz     4115€

But there is an incentive:
Until December 31, you get the 100Mpts for free, after all.

[nctnico]

I find it interesting the price difference between the two and four channel versions is so small. Like 5% to 10% of the price.

[Martin72]

I noticed that too and at first I thought maybe it was like the Rigol MSO5000, where the four channels are physically present but not fully activated.
But no, the 2-channel version of the HD3 is a “real” two-channel device.
That doesn't make it any better visually....An ugly thing.

[EEVblog]

I find it interesting the price difference between the two and four channel versions is so small. Like 5% to 10% of the price.

I suspect that the majority of 2CH sales go to educational institutions who aren't just buying onesies and twosies, so the prices you see wouldn't be reflective of what they get.

[EEVblog]

In my opinion, a new Keysight HD3 is not worth buying. There are too many spec downgrades, such as Waveform intensity grading, Sampling rate, etc. It is ridiculous to see the waveform intensity grading reduced from 256-level in MSO7104B -->  64-level in MSOX4104A --> 16-level in HD3.  I think Keysight gonna make a 4-level waveform grading intensity in a newer one. Geez, this is suck. The sampling rate is 3.2 Gsa/s? , downgraded from 4 Gsa/s. As of today, it should even be 5Gsa/s in this price range.  For those wanting to buy an HD3 at $10,000, getting a used EXR258A at $11,000 is better. Anyhow, EXR258A has a super long booting/start-up time.   I love previous Keysight products, but new products are not up to the standard versus the money spent.  https://www.keysight.com/used/us/en/oscilloscopes/exr258a-u08-2466971

As Mike said, 16 levels is a misprint, they increased it to 256.
And the sampling rate stays the same for all channels, it does not half with more channels on. And you have the remember it's 14bit vs 12bit ADC, so the sample rate is understandable.
I suspect that the complaints about the HD3 mostly all boil down to cost.

[Martin72]

The 14 bits are not an argument so far, at least not based on the examples we've seen so far, whether from you or from others.
I wouldn't worry about the “low” sample rate; for 14 bits, that's very OK and that they remain as they are, no matter how many channels are active, is something others could still learn from, keyword bandwidth reduction....
Personally, I don't like the way it looks and I don't like the small screen.
For €10,000 you get an MX04, which looks much better and has a larger screen.
I'm deliberately leaving out the 14 bits.
What I have to say in both cases (HD3, MX04), however, is that the build quality on the inside stands out clearly from other up-and-coming manufacturers....

[egonotto]

Hello,

if someone were to manipulate a 14-bit binary file from the HD3 in such a way that the two least significant bits are replaced by bits generated by a random number generator.

If this manipulation could be detected.

If so, how?

Best regards
egonotto

PS: For example, the signal measured for the 14-bit binary file is a sine wave at 500 MHz.

[Someone]

if someone were to manipulate a 14-bit binary file from the HD3 in such a way that the two least significant bits are replaced by bits generated by a random number generator.

If this manipulation could be detected.

If so, how?

If you said in advance that was the manipulation it would be easy to tell apart from the unmodified file as the correlation with surrounding data points would be destroyed while also adding quantisation noise to the retained signal.

[tycoon_9]

Thank you for the clarification. I saw a new datasheet to update to 256 level.

[tycoon_9]

In my opinion, a new Keysight HD3 is not worth buying. There are too many spec downgrades, such as Waveform intensity grading, Sampling rate, etc. It is ridiculous to see the waveform intensity grading reduced from 256-level in MSO7104B -->  64-level in MSOX4104A --> 16-level in HD3.  I think Keysight gonna make a 4-level waveform grading intensity in a newer one. Geez, this is suck. The sampling rate is 3.2 Gsa/s? , downgraded from 4 Gsa/s. As of today, it should even be 5Gsa/s in this price range.  For those wanting to buy an HD3 at $10,000, getting a used EXR258A at $11,000 is better. Anyhow, EXR258A has a super long booting/start-up time.   I love previous Keysight products, but new products are not up to the standard versus the money spent.  https://www.keysight.com/used/us/en/oscilloscopes/exr258a-u08-2466971

As Mike said, 16 levels is a misprint, they increased it to 256.
And the sampling rate stays the same for all channels, it does not half with more channels on. And you have the remember it's 14bit vs 12bit ADC, so the sample rate is understandable.
I suspect that the complaints about the HD3 mostly all boil down to cost.

Thank you for the clarification.  256-level trace intensity grading just makes me feel better about it.
The cost trade-off vs performance seems to be too stiff for buying justification, IMO ; where there are few keysight used MXR258A listed on market

[tycoon_9]

By the way, I am in the market for a new scope with a budget of around $10k—$15k. The caveat is that I have many Keysight HV diff probes, current probes, and a 1-GHz active probe. Comparing KS HD3, I like R&S MXO4, too, but I was locked down to KS due to my inventory of KS probes. 

[tycoon_9]

If you guys are working in Power electronics and have many HV diff probes, current probes, and some small signal active probes,==> you will find that switching oscilloscope brands is like moving from Apple iOS to Google Android and vice versa. 

[EEVblog]

By the way, I am in the market for a new scope with a budget of around $10k—$15k. The caveat is that I have many Keysight HV diff probes, current probes, and a 1-GHz active probe. Comparing KS HD3, I like R&S MXO4, too, but I was locked down to KS due to my inventory of KS probes. 

The test equipment equivalent of being locked into PCB software.

[MarkL]

If you guys are working in Power electronics and have many HV diff probes, current probes, and some small signal active probes,==> you will find that switching oscilloscope brands is like moving from Apple iOS to Google Android and vice versa. 

One approach is to standardize on Tektronix probes that have the TekProbe II interface.

Now that Tek's patent has expired, all the major scope manufacturers provide an adapter to use TekProbe II probes: LeCroy, Keysight, R&S, Rigol, Siglent, and even Tektronix themselves for their newer TekVPI scopes.  The adapters generally aren't cheap, but it beats re-buying the range of probes just because you want/need to switch scope manufacturers.

And for other manufacturers that don't offer an adapter, there is the vendor-agnostic Tek 1103 probe power supply that provides a BNC output from the probe.


Edit: added R&S

[egonotto]

By the way, I am in the market for a new scope with a budget of around $10k—$15k. The caveat is that I have many Keysight HV diff probes, current probes, and a 1-GHz active probe. Comparing KS HD3, I like R&S MXO4, too, but I was locked down to KS due to my inventory of KS probes. 

Hello,

Why is the used Keysight EXR258A not an option for you?

Best regards
egonotto

[tycoon_9]

By the way, I am in the market for a new scope with a budget of around $10k—$15k. The caveat is that I have many Keysight HV diff probes, current probes, and a 1-GHz active probe. Comparing KS HD3, I like R&S MXO4, too, but I was locked down to KS due to my inventory of KS probes. 

Hello,

Why is the used Keysight EXR258A not an option for you?

Best regards
egonotto

A used EXR258A is under my consideration.  One negative point ==> EXR258A is an MS window scope; the boot time to get ready for the scope screen is > 3 minutes.