New Rigol DS1054Z oscilloscope

[David Hess]

It would be as or more interesting to see the positive and negative transition time results under optimum conditions with the different bandwidth limits.  Then contrast those with the -3 dB bandwidth measurements to see how closely the oscilloscope follows (or does not follow) the 0.35 rule for a single pole roll off.

Then do the same at different vertical sensitivities and especially at the highest and lowest or at least when the high impedance attenuator is engaged.

[pa3bca]

Test with my 1074Z ("upgraded" to 100 MHz - Marmad is your scope still original 70 MHz? if so this might explain the 2 dB difference)
205 mV with no BWL (@ 100 MHz) and 73 mV with 20 MHz BWL = -8.97 dB

Tomorrow I will do some risetime tests with my Williams pulse generator

[pa3bca]

Test with a Williams pulse generator. 3 meters of coax on the collector to ensure a long enough pulse.

With no BWL the risetime is 3.1 ns. If we use 0.35 the Bandwidth would be 0.35/3.1e-9 = 113 MHz. Sounds plausible
With 20 MHz BWL the risetime is 14.7 ns, i.e. 0.35 / 14.7e-9 = 23.8 MHz.
Hmm looks ok'ish?

[marmad]

Marmad is your scope still original 70 MHz? if so this might explain the 2 dB difference

Yes, it's a loaner so it's unmodified. But the difference between our measurements was just slightly over 1dB.

[pa3bca]

Marmad is your scope still original 70 MHz? if so this might explain the 2 dB difference

Yes, it's a loaner so it's unmodified. But the difference between our measurements was just slightly over 1dB.

O yes. Well this 1 dB can be easily explained by the more limited BW of your scope so the amplitude is already a bit more down at 100 MHz than with my scope.
It looks like our scopes are probably not much different and seronday's  -11.3 dB is possibly faulty.

[marmad]

1) Is there any kind of information about these options?  I have found nothing beyond a brief sentence summary.

There's quite a bit of info in the user manual.

2) Are these options field-installable after the fact?  Or  must they be ordered with the unit?

They can be bought at any time (it's just a software key). There are also hacks available on this blog.

3) Does the "Serial Bus Analysis Option" possibly include Lan-C ?  (The Sony/Canon camera control protocol, see: EEVblog #297)
Or could I write a macro or something to add that to the protocols?

No - and no.

4) There is also an option: AT-DS1000Z - Advanced Triggering Option: RS232/UART,I2C,SPI,Runt,Windows,Nth Edge,
Do I need that ALSO in order to use the Serial Bus Analysis Option?

I don't think it's absolutely necessary (although I've never tried) - the Decode section has it's own definable parameters that are used irrespective of the trigger. But it makes life easier since you can get the Trigger working on the bus first, then just copy the Trigger settings to the Decode section.

[marmad]

O yes. Well this 1 dB can be easily explained by the more limited BW of your scope so the amplitude is already a bit more down at 100 MHz than with my scope.

That sounds about right. It's just a pity that the DS2000 has a better 20MHz limiter, even with double the base bandwidth, but I guess that's the drawback to the discrete front-end of the DS1000Z.

[miguelvp]

TEquipment is apparently out of stock, I guess that should be no great surprise
And I don't see any hint of when they think they might be in stock, probably not a big surprise, either.

I am very conflicted about when to order.  I am going to be out of town the first week of November,
and I don't want to order it now and possibly have them send it to be delivered while I am gone.
And I don't know how long the $399 price will remain?

He said 3 days ago that they have 100 more in order coming this week within 5 days so any time now and that most are sold out but some might be left, then another order 4 weeks later, why don't you PM Evan?
https://www.eevblog.com/forum/chat/special-price-for-eevblog-members/msg535230/#msg535230

Edit: as for your other questions, they do come with a free trial, so you can see if they fit your need before purchasing if you do feel uncomfortable by "unlocking" them like some are doing.

If you purchase them or unlock them, it's a serial code that you enter with the scope after the fact.

[miguelvp]

3) Does the "Serial Bus Analysis Option" possibly include Lan-C ?  (The Sony/Canon camera control protocol, see: EEVblog #297)
Or could I write a macro or something to add that to the protocols?

No - and no.

Not really an absolute No, he could take the recorded waveform and convert it to digital levels respecting the timing and feed that to his own decoder.

[seronday]

In seronday's graph, he shows 100MHz as being attenuated to about -11.3dB when the 20MHz BW limit is turned on. My test shows a 100MHz 400mV sine being attenuated to ~-7.8dB.

Marmad,
The reference frequency for all levels on the Graph in question is 1Mhz.
You appear to be measuring the difference in levels at 100Mhz which is in fact approx. 8dB.

[marmad]

3) Does the "Serial Bus Analysis Option" possibly include Lan-C ?  (The Sony/Canon camera control protocol, see: EEVblog #297)
Or could I write a macro or something to add that to the protocols?

No - and no.

Not really an absolute No, he could take the recorded waveform and convert it to digital levels respecting the timing and feed that to his own decoder.

That goes without saying; you can do the exact same thing for any DSO that can save a waveform (virtually every one manufactured). But that is neither a macro nor something you can add to the existing protocols (i.e. internally).

[marmad]

Marmad,
The reference frequency for all levels on the Graph in question is 1Mhz.
You appear to be measuring the difference in levels at 100Mhz which is in fact approx. 8dB.

Ok. But I didn't see that mentioned in your original post (although I now see the graph begins at 1MHz) - and why 1MHz for a 100MHz DSO?

[seronday]

Marmad.
When measuring the frequency response of an instrument such as an oscilloscope you would normally start at DC and work upwards in frequency from there, to find and record any variations in level.

A quick check doing this on a DS1074Z shows that the frequency response is essentially flat up to about 12Mhz and then starts to roll off.
When the 20Mhz Band Limit Filter is selected the roll off starts at approx 3Mhz.

Generally in the performance verification documentation for an oscilloscope, the manufacturer specifies the reference frequency to use for the bandwith performance measurement.
This document does not appear to exist as yet for the Rigol DS1000Z series.
There is however such a document for the DS4000 series in which they do specify using 1Mhz as the reference frequency.

[David Hess]

Test with a Williams pulse generator. 3 meters of coax on the collector to ensure a long enough pulse.

With no BWL the risetime is 3.1 ns. If we use 0.35 the Bandwidth would be 0.35/3.1e-9 = 113 MHz. Sounds plausible
With 20 MHz BWL the risetime is 14.7 ns, i.e. 0.35 / 14.7e-9 = 23.8 MHz.
Hmm looks ok'ish?

They look fine to me.  The 3.1 nanosecond example might be showing dribble-up in the charge line.  I wonder what that little dip is before the leading edge in both examples.  I have seen it before and been given several explanations depending on the oscilloscope design.

I was chatting in email with someone about the design and came up with a calculated value of below 28 MHz for the 20 MHz bandwidth limit from Dave's reverse engineered schematic.

[David Hess]

Marmad,
The reference frequency for all levels on the Graph in question is 1Mhz.
You appear to be measuring the difference in levels at 100Mhz which is in fact approx. 8dB.

Ok. But I didn't see that mentioned in your original post (although I now see the graph begins at 1MHz) - and why 1MHz for a 100MHz DSO?

The reference level on an SG503 leveled sine wave generator intended for testing oscilloscopes up to 250 MHz is 50 kHz.  On an SG504 intended for testing from 250 MHz to 1 GHz, it is 50 kHz and 6 MHz.  The flatness of the leveled generator will be specified in comparison to the reference frequency and they will perform better than typical RF test equipment; amplitude variation could be within 4% (0.35dB) or better.  The graph I posted in that discussion thread from a 2232 should be within +1.5/-1.0% or about 0.22 dB peak to peak error.

Calibration of a leveled sine wave generator itself presents an interesting dilemma.

Like seronday says, the oscilloscope calibration instructions or specifications often or usually include the reference frequency used for flatness.

[marmad]

Marmad.
When measuring the frequency response of an instrument such as an oscilloscope you would normally start at DC and work upwards in frequency from there, to find and record any variations in level.

Doh!   Wow, somehow my left brain was malfunctioning yesterday (probably from breathing too much dust - I'm in the middle of a demolition of my old kitchen) and I somehow overlooked the obvious. Sorry about that (although I still don't understand why your frequency axis is non-uniform  )!

[ataradov]

It looks like shipping date from Tequipment has moved to November 20, 2014

[ataradov]

False alarm

You may have received an automated message stating your ship date was changed to 11/20/2014.  Please accept our apologies for any inconvenience, as your order is firm and your ship date is ~11/3/2014. 

[nkw]

False alarm

You may have received an automated message stating your ship date was changed to 11/20/2014.  Please accept our apologies for any inconvenience, as your order is firm and your ship date is ~11/3/2014. 

Can I ask how/when you received the second message? I received the delay e-mail with the 11/20 date this morning but not the second message with the 11/3 date. I'm trying to decide if I should cancel with tequipment and look someplace else.

Thanks.

[ataradov]

It came from Christine Hoh from Tequipment  35 minutes ago.

I'd contact them directly if you have doubts.

[nkw]

It came from Christine Hoh from Tequipment  35 minutes ago.

Received an email from them about 15 minutes after I posted about it. Nov-3 is much better than Nov-20. Thanks!

[tequipment]

Just to update everyone.  We have 85 units coming in on this shipment.  This shipment is sold though.  If you are part of the 85 that are inbound, you got an email about the error in pushing the date back. 

We have a special air shipment of 100 units coming Nov 20th and then we have a  bunch more arriving in early December.  We have some units on the November shipment that are not sold yet. We have the only stock in the USA at this point until December because of our big pre orders.

Thanks for the business!
Evan Cirelli
TEquipment.NET

[JoeO]

Thanks for the update Evan. 

I have also received the 2 emails with the 11/3 ship date in the second email.

JOe

[Micael]

Got this email from batronix:

"On 24th October you have ordered a Rigol DS1054Z.

Herewith, we would like to inform you that your requested device is not available yet.

At the moment, we are waiting on three deliveries from Rigol for this device. The estimated time of arrival for the first delivery lies between November 7th - 10th. The second delivery approximately at the beginning of December, the third around mid of December.

Due to your position in our waiting list, you will receive a device from the second delivery. We are sorry for the inconvenience.

We are apologising for this delay. The demand for this devise is unexpectedly high. Rigol itself can currently not manufacture as many devises as needed at the moment.

If the time of waiting is too long for you, you are welcome to choose a similar product from this series. We grant you a 8% discount because of the special circumstances.

Best regards,

Thorsten Schliszio"

I guess the DS1054Z is going to be an early Christmas present over here.

[argg]

same here,  I ordered today and they promised to ship with the delivery they expect on Dec 6.  Early Xmas for us indeed