DSO oscilloscope saving single shot trigger event to memory for power supply startup

Go To Last Post
28 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I need a feauture for a DSO, but dont know which DSO brands are capable of doing it. My Fluke 199C seems not to have this feuture.

I want to save the first 100ms of the power supply startup sequence to DSO memory. 

I want to set a triger event ex: If the input goes over 100mV, the DSO will start to save the whole sequence until the memory is full. And very important if the memory is full, it will not roll to the beginning of the memory and rewrite the memory. It will just stop after the memory is full. It should be possible to zoom and look at this long sequence. 

Example:

DSO have 1Mpoint memory. 

Time division is 10ns.

My event starting trigger level is set as 100mV. It should mean if the input signal rises above 100mV, DSO will start to save with a sampling rate of 10ns until the memory of 1Mpoints is full. This means with a sampling interval of 10ns , 1Mpoints will be full after 10ms. 

Once the starting event trigger is satisfied, the DSO should go ahead with the saving sequence independent of the trigger level.

Can any DSO do that?. If yes which one?

(I am trying to save the startup of a power supply as detiled as possible). Each point of the output with a resolution of 10ns after the input raises above 100mV.

Best regards.

________________________________ We dream of a world where current does not need the voltage to flow.

Last Edited: Wed. Mar 20, 2019 - 11:19 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

incal wrote:
... to DSO memory.
then to LAN?

Some DSO have logging software on a PC.

incal wrote:
DSO have 1Mpoint memory.
and 1Gbps Ethernet :

InfiniiVision 1000 X-Series Oscilloscopes | Keysight (formerly Agilent’s Electronic Measurement)

Oscilloscope Control & Automation | Keysight

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Good idea, Innovision 1000 X is also an economic option.

 

If we sample every 10ns, it means 100Msamples/s.  If each sample is 10 bit. it means 1Gbit/second data flow.  125Mb/s.

 

Do you think that this DSO is capable of extracting this data flow (125mB/s) over 1 gbit ethernet to an external hard drive.

 

1 gbit ethernet can handle it on limits. Any SSD should be able to save it.

 

But I am not sure if the DSO can process this amount of data and send it to the LAN?  

 

Any how , I believe if we reduce the resolution (which means to sacrifice of the analyse of gate rise times of semicondutor switches) 

 

It may be an option. But for 10ns sampling rate, I am a little bit skeptic.

________________________________ We dream of a world where current does not need the voltage to flow.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

incal wrote:
Do you think that this DSO is capable of extracting this data flow (125mB/s) over 1 gbit ethernet to an external hard drive.
A question for a Keysight FAE or a distributor FAE as I have no answer.

incal wrote:
But I am not sure if the DSO can process this amount of data and send it to the LAN?
Am sure it can as these likely have the internal oscilloscope app on an RTOS (ADC -> S/H -> ASIC -> DMA into DRAM -> app on RTOS on CPU); Agilent's "old" 2.5GHz scope ran Windows XP and it has the throughput.

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks for the info. I should ask the local distrubutor if it is possible.

 

If it is possible there is no need for trigger management. You start to record , turn the power supply on and capture the whole data to an external hard drive.

Even a fast USB drive that is available on the front panel of the agilent DSOX1102A can be enough if it is possible to save real time data to it. 

 

There are ssd based usb drives.

 

Thank you for the idea. 

________________________________ We dream of a world where current does not need the voltage to flow.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Unless I've missed something in #1, the trigger mode you've described is quite a basic level trigger mode and I'm almost certain Scopemeter supports it. In fact I haven't seen a DSO that doesn't support it. Think many years back to Tektronix 2220  . . . Yes I believe it did.

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

N.Winterbottom wrote:

Unless I've missed something in #1, the trigger mode you've described is quite a basic level trigger mode and I'm almost certain Scopemeter supports it. In fact I haven't seen a DSO that doesn't support it. Think many years back to Tektronix 2220  . . . Yes I believe it did.

 

 

 

The problem with Fluke 199C scopemeter is:

 

It has 2 options. 

 

- Single shot trigger: It has 3000 points of storage. So the resolution is very bad. I must arrange 15-20 seperate windows and correspanding trigger levels to get a picture. Even then the picture is not so accurate. 

 

- Scope record: 27500 points of storage. But the whole window can not be smaller than 6 second?!?. and the minimum time division is unfortunatelly 5ms. I can digitally zoom (I think it is an extrapolation) this 5ms 1:10 and what I get is 500us time divisions. But as they are extrapolated from 5ms data, they are very blurry.

 

I did not have another DSO (only old analog). So I wanted to know if there are better choices. 

 

I found a cheap oscilloscope from AAtech ADS-3072B.  It has 2Mpoints of storage. It means with a single shot trigger If my whole screen is 20 ms, I can zoom the details up to 10ns. This DSO costs only 300 USD. 

 

I found also another model  AAtech ADS-4012. It has 14Mpoints. If my whole screen is 280ms, I can zoom into details up to 10ns. this is very attractive for me.

 

The last one I found is Rigol MSO5074 4 channels (costs 999 usd) It has 200Mpoints !!!!!! For a screen size of 2000ms (2 seconds) I can zoom into 10ns details. 

 

Would you recomnend it? 

________________________________ We dream of a world where current does not need the voltage to flow.

Last Edited: Wed. Mar 20, 2019 - 02:41 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'm with Witnerbottom here, From what I understand what you describe is one of the basic features of any decent scope.

I see he 199C is some kind of hand held thingie:

https://duckduckgo.com/?q=Fluke+199C&iax=images&ia=images

Maybe it does not have Ethernet at all?

 

Does the connectivity have to be via Ethernet?

Ethernet on a socope is pretty common nowaday's, but some of the older or simpler digital scopes only have USB (slave) or USB (Flash drive) for storing data.

 

What sort of budget are you thinking about?

Rigol DS1054Z has been sort of the defacto default for a pretty decent scope for a low budget. (EUR400)

In the last year or so the Siglent SDS1104X-E (EUR500) seems to have become more popular than the Rigol.

 

The Siglent has an integrated Web interface and can be acessed with any regular web browser, but at least when this scope hit the market this was uncomfortable slow (Multiple seconds delay to react to a mouse click). The Rigol was pretty buggy when it was first launched, but it appears that all the bugs have been ironed out a few years ago.

 

These are both "low budget" scopes in today's market, but perform better than mid-range scopes of 10 years ago. The Scope market has made Giant leaps, and stuff like Ethernet is (almost?) standard.

 

If you have a budget of EUR 1000 to EUR 2000 or more then all the traditional brands (Tektronics, Keysight,  Rohde&Sqartz, etc) come into view.

These probably have a more "fluent" user interface and less bugs. There also (finally) seems to be a slow trend of higher resulution ADC's and some of even the low budget scopes have 10 or even 12-bit ADC's. 

 

incal wrote:
Do you think that this DSO is capable of extracting this data flow (125mB/s) over 1 gbit ethernet to an external hard drive.

You really do not want a scope with a communication bandwith of 125mB/s. You would need 3 months to transfer a Mega Byte:

>>> 1/.125
8.0
>>> _*1e6
8000000.0
>>> _/3600
2222.222222222222  # Hours.
>>> _/24
92.5925925925926   # Days.

But your whole question about bandwith is silly anyways.

You want single trigger mode, and the data transfer has nothing to do with the capture speed.

Doing magic with a USD 7 Logic Analyser: https://www.avrfreaks.net/comment/2421756#comment-2421756

Bunch of old projects with AVR's: http://www.hoevendesign.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

 but dont know which DSO brands are capable of doing it.

 

The hand held instruments are great if they meet your needs...

 

But, as already mentioned above, ANY decent DSO will have a "Single Shot" mode, which once triggered will fill its memory, and then allow you to zoom in and scroll back and forth through the data.

 

I have a Rigol DS1062CA, (2 Ch, 60 MHz, 2GSa/sec), so a decent hobbyist level Oscope, but certainly not in the professional grade.

It does this just fine. 

It is an older model that has been superseded by newer better equipment.

 

JC

 

Edit:Typo

 

 

Last Edited: Wed. Mar 20, 2019 - 03:03 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

But your whole question about bandwith is silly anyways.

You want single trigger mode, and the data transfer has nothing to do with the capture speed.

 

In power supply design , It would be a great comfort for me being able to analyse maybe 8-10 seconds of data with 10ns details.

 

Dynamic load change, detailed startup inrush and different scenarios can be comfortly analysed with seconds of storage. But such a device having 1Gpoints of storage is over 10.000 USD.

 

A very fast interface (if possible) would allow a direct transfer of this data without a need for deep memory.

 

This was the idea behind gbit ethernet interface.

 

Theoretically if such a connection can be established through an RTOS kernel of the osciloscope, you can record 100s of seconds (till your ssd is full) within ns seconds of resolution. And if it is possible you can do it with a much cheaper device.

 

 

________________________________ We dream of a world where current does not need the voltage to flow.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

incal wrote:
In power supply design , It would be a great comfort for me being able to analyse maybe 8-10 seconds of data with 10ns details.
Time domain and frequency domain are linked.

Testing power supply: Measuring stability - Power House - Blogs - TI E2E Community

Easy Control Loop Measurement and Verification - ON Semiconductor

 

Dynamic load change, ...

can be linked to frequency domain though that's another tool (an oscilloscope is easier)

 


Using the Network Analyzer [Reference.Digilentinc]

The Electrical Engineer’s Multi-tool | Keysight Community (frequency response analysis)

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

DocJC wrote:

 but dont know which DSO brands are capable of doing it.

 

The hand held instruments are great if they meet your needs...

 

But, as already mentioned above, ANY decent DSO will have a "Single Shot" mode, which once triggered will fill its memory, and then allow you to zoom in and scroll back and forth through the data.

 

I have a Rigol DS1062CA, (2 Ch, 60 MHz, 2GSa/sec), so a decent hobbyist level Oscope, but certainly not in the professional grade.

It does this just fine. 

It is an older model that has been superseded by newer better equipment.

 

JC

 

Edit:Typo

 

 

 

I know that this is a standart feauture of a DSO. But what I wanted to mention is the duration of the storage time for a resolution of 10 ns (This is very common for high frequency power supplies with SiC and GaN mosfets where the dead times can be as small as 50 ns and the gate drive rise times can be as low as 5ns).

 

Yes it may be a standard feature (single shot trigger) . But for a duration of seconds with high resolution you should invest a lot of money.  Especially for testing some dynamic loads , it may be a headache to arrange trigger levels and find the suitable frames. Typically a dynamic load test duration is from 100ms to 2000ms.  It is very effective to be able to see the whole sequence in one picture.

 

 

________________________________ We dream of a world where current does not need the voltage to flow.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

gchapman wrote:

incal wrote:
In power supply design , It would be a great comfort for me being able to analyse maybe 8-10 seconds of data with 10ns details.
Time domain and frequency domain are linked.

Testing power supply: Measuring stability - Power House - Blogs - TI E2E Community

Easy Control Loop Measurement and Verification - ON Semiconductor

 

Dynamic load change, ...

can be linked to frequency domain though that's another tool (an oscilloscope is easier)

 


Using the Network Analyzer [Reference.Digilentinc]

The Electrical Engineer’s Multi-tool | Keysight Community (frequency response analysis)

 

 

Yes . Root locus compensation is essential. But I prefer doing it in time domain. Because it is easier. By trial and error I get good results. 

 

In frequency domian you need more mathematical work and I feel myself no so comfortable in this region. I know how to compensate the loop but again i enyoj doing it in the time domain. Especially for resonant power supplies the variation of dead times in ns intervals respect to different load conditions have great effects to efficiency and control stability. 

Having a picture of maybe 10-20 seconds of a sequence with ns resolution will be just like driving a ferrari for me.  

________________________________ We dream of a world where current does not need the voltage to flow.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

When I had a (employers) Scopemeter I must confess I often ran into the problem of low acquisition memory. The "Scope Record" feature is squarely aimed at long term "monitoring" over durations of minutes or even hours. As such it doesn't help with your capture scenario.

 

My workaround for when I ran out of memory was to leave the trigger level alone but successively increase the trigger delay so I could piece-by-piece, examine the event in reasonable detail.

 

I was eventually allowed to upgrade, but because Tektronix was a sister company, it was mandated I choose a scope from their range. Oh dear oh dear.

 

I chose a Tektronix MSO2014 and only missed the fully isolated channels from the old Scopemeter.

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I get it, the sample depth is costly.

 

Saelig has a Rigol, 4 CH, 500 Mpts, for $10K.

And if you are spending that much, then obviously you have to look at some other high end scope manufacturers, also.

 

If the company is going to sell lots of these power supplies, then perhaps having the correct design equipment is simply the cost of doing business.

 

Alternatively, perhaps you could rent a high performance O'scope for a few weeks?

 

JC

Last Edited: Thu. Mar 21, 2019 - 12:33 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

DocJC wrote:
I get it, the sample depth is costly.
Scope yes, logic analyzer no (USB 3, DMA into PC's DDR DRAM)

Analyzer2Go - Turn your development board into a logic analyzer

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

N.Winterbottom wrote:

When I had a (employers) Scopemeter I must confess I often ran into the problem of low acquisition memory. The "Scope Record" feature is squarely aimed at long term "monitoring" over durations of minutes or even hours. As such it doesn't help with your capture scenario.

 

My workaround for when I ran out of memory was to leave the trigger level alone but successively increase the trigger delay so I could piece-by-piece, examine the event in reasonable detail.

 

I was eventually allowed to upgrade, but because Tektronix was a sister company, it was mandated I choose a scope from their range. Oh dear oh dear.

 

I chose a Tektronix MSO2014 and only missed the fully isolated channels from the old Scopemeter.

 

 

This is exactly what I am doing now and it can be a headache when dealing with dynamic loads where a non seperable time window is a great advantage. For a dynamic load it is not possible to analyse seperate time frames one after another and piece by piece. The sequence should be recorded at once.

 

The main problem as I told is the isolated inputs. 1 1000V Cat III isolated input probe costs up to double the price of a good DSO.

 

In between the new models of scopemeter have  a memory of 10.000 points. But I have an old one with 3000 points. 

 

________________________________ We dream of a world where current does not need the voltage to flow.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

DocJC wrote:

I get it, the sample depth is costly.

 

Saelig has a Rigol, 4 CH, 500 Mpts, for $10K.

And if you are spending that much, then obviously you have to look at some other high end scope manufacturers, also.

 

If the company is going to sell lots of these power supplies, then perhaps having the correct design equipment is simply the cost of doing business.

 

Alternatively, perhaps you could rent a high performance O'scope for a few weeks?

 

JC

 

I am very suprised that 2 isolated channels on my Fluke scopemeter 199C , 1000V cat III cost over 3000 USD if you want to buy them for a regular DSO. So if I rent an osciloscope to see the difference I should also rent these probes. It is also unlikely that these probes are available for rent.

 

The problem with these probes: There are cheaper ones (600-700 USD). But they have a frequency degradation. They are ok up to 3-4Khz. But the insulation rating goes down to 100V for 20kHZ for a cheap one.

________________________________ We dream of a world where current does not need the voltage to flow.

Last Edited: Thu. Mar 21, 2019 - 09:46 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

How good a measurement are you interested in taking?

 

What is the bandwidth of the probe required to pass any glitches you wish to see?

Voltage sags don't require a great BW, but spikes, (from anywhere), obviously do.

Overall rate of rise under various loads is again a low BW requirement.

Power Supply oscillation monitoring is also likely not a very high frequency requirement.

 

You could spend a little time seeing just how hard it is too make your own HV probes.

Obviously not super fancy and nicely encased, but perhaps functional to meet your overall purposes.

 

Next, your either on board with this idea or you are going to think its crazy, build your own data sampler.

I'm sure the guys who use programmable arrays would do this differently, but I'm old school.

 

Set up 10 micros, (or whatever), and have each one dump its data in a circular buffer, then to its own FTDI USB chip.

Then have one master Micro, or clock and low level logic, (counter chips), which triggers each micro and its own ADC in succession.

The data acquisition rate for any one micro is only 1/10th the over sampling rate.

You can put a cheap flash 8-Bit ADC converter on the front end of each of the micros.

The FTDI parallel input to USB chip has a much higher BW than the serial input to USB FTDI chip that is usually seen in projects here on the Forum.

10 of those gives you quite a large data path to a high speed PC.

The PC can both reconstruct and store the original signal, easily.

The single HV probe, (purchased or built by you), feeds all 10 ADC inputs with the same signal.

The ADC's and the micros just sample in a round robin manner.

You may or may not require an op-amp or two to voltage clamp and buffer the input signal from your HV probe.

One would have to look at USB chip options to maximize your data rates, and the ability to the PC to simultaneously handle multiple channels without data loss.

 

I built a much lower BW device similar to the above with 5 micros, four for signal sampling in a round robin fashion and one for the master control and clock generation, for measuring 0 to +/- 2KV or so defibrillator signals long ago.  I had only one data path to the PC.  In any event, the project is at least feasible.

 

Now decide the value of your time to build your own data acquisition system, vs. renting / purchasing the required equipment and getting along with your primary mission.

 

JC

 

Edit:  This project ended up on a PCB, but all I can find a photo of is the prototype.

One photo is the "HV Probe".

The other is the five (PIC) micros, four for sequential data sampling, the fifth was the controller.

The mini-FTDI USB breakout PBC is also visible.

 

 

 

 

 

 

Last Edited: Thu. Mar 21, 2019 - 11:19 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Observation:

 

For power supply startup monitoring, you DON'T need a fast sample rate, at all. A few KHz ought  to be fine. At that rate, memory requirements are manageable and data can be sent, real-time via USB or  WiFi.

 

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ka7ehk wrote:

Observation:

 

For power supply startup monitoring, you DON'T need a fast sample rate, at all. A few KHz ought  to be fine. At that rate, memory requirements are manageable and data can be sent, real-time via USB or  WiFi.

 

Jim

 

Assuming I have a cheap DSO with a fast connection to PC. (ex : 1gbit ethernet). Do I need the internal deep memory in this case if I want to transfer the samples in real time to a pc for recording. 

This is the question I can not answer.

 

For example cheapest USB oscilloscopes  has USB3 connection to PC. But they may have poor internal memory?. Can they stream the data in real time to pc?. (I asked the same question on USB picoscope forum, no one answered) 

________________________________ We dream of a world where current does not need the voltage to flow.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

DocJC wrote:

How good a measurement are you interested in taking?

 

What is the bandwidth of the probe required to pass any glitches you wish to see?

Voltage sags don't require a great BW, but spikes, (from anywhere), obviously do.

Overall rate of rise under various loads is again a low BW requirement.

Power Supply oscillation monitoring is also likely not a very high frequency requirement.

 

I want to make some observations on the new SiC and GaN mosfets. They have gate signals rising in 10ns and bridge dead times as low as 100ns. I believe my scopemeter with 3000 points may be on the limit with some headache with regular semiconductors up to 15-20kHZ. 

 

But the trend go to the diretion of SiC and GaN mosfets. It is very common to have 500kHz-1Mhz switching frequencies for this technology. This enviroment will need more precision test and measuring equipment.

 

Rigol DSO has 300V Cat II inputs. (it means 400Vpeak). 

 

In worst case I may construct a voltage divider with low inductance (using smd HV resistors) after the probe . But such a circuit will have a very bad compensation. The pF capacitor on the probe will interact with the high ohmic voltage divider and rise and fall times of the the signal on the probe will be very inaccurate.  

 

What happens when I insert a isolation transformer (230Vac to 230Vac) on the input of the DSO and use a high ohmic voltage divider after the standart 1Mohm/10pF probe.  For example 1:10 or 1:5   1Mohm:5MOhm just between the porbe and measuring point?

 

How can I compensate the rise and fall times of modified probe . I believe the adjustment screw on the probe will not be enough if I add 1Mohm:5mOhm voltage divider after the probe.

 

What dou you think abaout a combination of isolation transformer at the supply input of the DSO and a voltage divider after the probe?. How should I compansate the probe? 

 

 

 

 

 

 

 

________________________________ We dream of a world where current does not need the voltage to flow.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I want to make some observations on the new SiC and GaN mosfets. They have gate signals rising in 10ns

To a certain extent, so what?

Logic gates can easily have risetimes of 5ns, yet I don't need to grab a 2GHz scope to look at my digital alarm clock, even a 20MHz scope is plenty. 

What if you installed some super-high speed buffers that could drive your gates in 30ps---would you need a 10GHz scope for a 1 MHz switcher...nope

 

Be creative in your testing...a few hundred milliseconds of runtime should capture any worthwhile event.  You might need to run different tests for different events, but that's what extensive testing (lots of tests) is all about.

Since you are presumably in control of the test (rather then some once a year random event), you can control the time range. 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

avrcandies wrote:

I want to make some observations on the new SiC and GaN mosfets. They have gate signals rising in 10ns

To a certain extent, so what?

Logic gates can easily have risetimes of 5ns, yet I don't need to grab a 2GHz scope to look at my digital alarm clock, even a 20MHz scope is plenty. 

What if you installed some super-high speed buffers that could drive your gates in 30ps---would you need a 10GHz scope for a 1 MHz switcher...nope

 

Be creative in your testing...a few hundred milliseconds of runtime should capture any worthwhile event.  You might need to run different tests for different events, but that's what extensive testing (lots of tests) is all about.

Since you are presumably in control of the test (rather then some once a year random event), you can control the time range. 

 

I also do not think and say that I need a high bandwidth scope. 50Mhz-70Mhz should be enough even for high frequency power electronic.

But it would be nice if I had a little more memory to see whats happening for the test scenario of a dynamic load change. Especially for topologies such as LLC resonant converters. 

________________________________ We dream of a world where current does not need the voltage to flow.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If you want fast logging to a PC then maybe one of the PC based instruments is more to your liking.

Picoscope is a well known brand.

There's also "Cleverscope" https://cleverscope.com/products/

Some of the clever scope devices have isolated inputs and a pretty high bandwidth for streaming to a PC.

The isolation is done in the digital domain, not in the analog front end.

Doing magic with a USD 7 Logic Analyser: https://www.avrfreaks.net/comment/2421756#comment-2421756

Bunch of old projects with AVR's: http://www.hoevendesign.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Paulvdh wrote:
There's also "Cleverscope" https://cleverscope.com/products/
Digilent Analog Discovery 2 has the same ADC spec as Cleverscope 300-series though a third the bandwidth.

 

Analog Discovery 2 [Reference.Digilentinc]

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Cool instrument!

DocJC wrote:
I'm sure the guys who use programmable arrays would do this differently, but I'm old school.
Typically, a FPGA tool suite will have schematic capture for design entry so don't need to learn an HDL (a sampler is straight forward)

Seems AFE are somewhat regular new arrivals at the distributors; so, impedance match between probe tip to AFE input.

 

AFE - Analog Front End

https://www.cypress.com/products/universal-serial-bus-usb#tab3

[GPIF]

Overview :: EZUSB communication core :: OpenCores

 


edit : Welcome to icestudio’s documentation! — Icestudio 0.3.3-rc documentation

via FOSDEM 2018 - Graphic design tools for Open Source FPGAs

 

"Dare to be naïve." - Buckminster Fuller

Last Edited: Fri. Mar 22, 2019 - 10:56 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

DocJC wrote:
Saelig has a Rigol, 4 CH, 500 Mpts, for $10K.
A new Rigol series has 100Mpts with a 200Mpts option; starts at 70MHz bandwidth, no FRA.

MSO5000 Mixed Signal Oscilloscopes and our Best Digital Oscilloscope  | RIGOL

UltraPowerAnalyzer Data Sheet.pdf (a software option)

 

"Dare to be naïve." - Buckminster Fuller