ARM Architecture

Oscilloscope assessment: Hameg HMO2024 – Jason Sachs


Final yr I wrote about a few of the key traits of oscilloscopes which can be vital to me for working with embedded microcontrollers. In that weblog entry I rated the Agilent MSOX3024A 4-channel 16-digital-input oscilloscope extremely.

Since then I’ve moved to a distinct profession, and I’m once more looking out for an oscilloscope. I nonetheless think about the Agilent MSOX3024A the only option for a general-purpose benchtop oscilloscope — at roughly US$5400, it is a pretty good worth, though nonetheless not cheap.

Whereas in search of different another check tools, I not too long ago ran throughout the web site of Hameg, which has a sequence of mixed-signal oscilloscopes. Hameg is a German firm that is a subsidiary of Rohde & Schwarz; each corporations are pretty well-known in Europe however a lot much less acquainted right here within the USA. The Hameg HMO2024 is a 200MHz, 4-channel analog, 8-digital-input benchtop oscilloscope that retails for roughly US$3000, or 45% lower than the MSOX3024.

I not too long ago had an opportunity to check out the HMO2024 and see the way it compares to the Agilent scopes I’ve tried. Here is what I discovered.

Simply to recap, there are a selection of vital options I think about important:

  • 4-channel analog
  • Digital scope inputs
  • Moveable w/ minimal footprint
  • “Hello-res” acquisition mode (boxcar filter to common a number of samples for elevated efficient decision)
  • Straightforward switch of screenshots and waveform information to a PC (ideally by way of both USB “thumb drive” or by way of ethernet)
  • Intuitive consumer interface
  • Choices to decode and set off on I2C/SPI/UART/CAN/LIN

Apart from the consumer interface, all of those options/choices are described within the specs of the Hameg HMO2024, and it regarded like a sexy different to the Agilent scope for engineers with a extra restricted finances.

Analog traits

The HMO2024 has 4 analog channels. The vertical acquire and offset are managed by a single pair of knobs, making use of to the currently-selected channel, relatively than a devoted vertical acquire and offset knob per channel. At first I used to be involved that this may change into inconvenient and cumbersome, however I discovered the flexibility to pick out a channel and alter the acquire and offset very straightforward to make use of.

The vertical offset of the analog channels is pretty versatile and happens previous to the oscilloscope ADC: you may offset a channel by as much as ±10 divisions, so in case you’re a 2V/division waveform, you may offset it by as much as ±20V. That is significantly better than the lower-end Tektronix oscilloscopes (MSO2000, MSO3000 sequence) the place the offset occurs post-ADC and also you simply get the flexibility to maneuver the scope hint up and down, however not so good as the Agilent habits.

For comparability I ran some checks on the HMO2024 and an Agilent MSO6014A oscilloscope that we now have at our firm, 2 channels: one with the scope probe and floor enter tied collectively to chassis floor, and the 2nd probe related to a 24V energy provide. The Agilent MSO6014A was in a position to offset the 24V sign so it confirmed up on the display even at 10mV/division: a powerful 2400 divisions of offset!

Here is some statistics on noise:
















Take a look at RMS 99%-1% span peak-to-peak  
HMO2024, 0V enter, 1msec/div, sampling mode 4.7mV 15.6mV 31.6mV  
HMO2024, 24V enter, 1msec/div, sampling mode 91mV 320mV 480mV  
HMO2024, 0V enter, 1msec/div, hi-res mode 1.4mV 6.5mV 15.1mV  
HMO2024, 24V enter, 1msec/div, hi-res mode 20mV 80mV 160mV  
HMO2024, 0V enter, 10usec/div, hi-res mode 1.7mV 7.8mV 26.2mV  
HMO2024, 24V enter, 10usec/div, hi-res mode 26mV 120mV 180mV  
MSO6014A, 0V enter, 1msec/div, 10mV/div sampling mode 3.0mV 13.8mV 23.8mV  
MSO6014A, 24V enter, 1msec/div, 10mV/div sampling mode 3.6mV 13.8mV 23.8mV  
MSO6014A, 0V enter, 1msec/div, 10mV/div hires mode 0.5mV 2.2mV 4.0mV  
MSO6014A, 24V enter, 1msec/div, 10mV/div hires mode 0.7mV 3.3mV 13.1mV  
MSO6014A, 0V enter, 10usec/div, 10mV/div hires mode 3.9mV 17.5mV 28.8mV  
MSO6014A, 24V enter, 10usec/div, 10mV/div hires mode 3.5mV 16.3mV 37.5mV  

The HMO2024 has noise traits on par with the MSO6014A for the 0V enter. For the 24V enter the MSO6014A had an enormous benefit over the HMO2024 as a result of the acquire may be elevated with a big offset to maintain the scope waveforms inside the seen space.

Triggering choices are fairly intuitive, and there are plenty of choices: slope set off, twin slope set off (known as “B-trigger” of their handbook, which is a sign exceeding a threshold after which later dropping under one other threshold, with set off rely or time {qualifications}), pulse set off, logical mixtures of states (good for the digital inputs), video, or serial bus set off.

There may be additionally a devoted AUTO/NORM key, which I discovered very useful. What number of occasions have you ever run into this case: you get your set off situations outlined, and are all able to go along with Regular triggering, and what do you see in your scope? Nothing. Darn. So that you surprise what is going on on, and the one approach to see the oscilloscope inputs is to go to Auto triggering. Then you need to modify your settings and return to Regular triggering. Repeat as obligatory. Effectively, with most scopes this requires a number of button/menu alternatives, and it shortly turns into a ache. With this scope it is a single button press, which makes this a snap.

Most oscilloscopes have a set off holdoff setting you may modify, that inhibits subsequent triggers for a sure interval after a set off has occurred. The HMO2024 doesn’t have this characteristic. I am used to having it, so I am somewhat dissatisfied by its absence — it is helpful when you’re triggering on lengthy sequences with areas in between, so as to have the scope disregard triggers that may occur in the course of a sign. I suppose typically you may accomplish the identical factor with a pulse set off, however one way or the other it looks like there are circumstances the place a set off holdoff could be vital. 

This scope has a Zoom characteristic to amass lengthy waveforms after which view a smaller time interval inside the acquired waveform, whereas viewing your entire waveform in a small portion of the display. That is additionally very helpful, and places the HMO2024 consistent with different scopes with this characteristic just like the MSOX3000 sequence.

There are additionally the same old cursor measurement and automeasurement options to compute imply/max/min/peak/frequency/and many others. values. The measurement characteristic had one thing I used to make use of a very long time in the past after I had entry to one of many LeCroy scopes: a “statistics” show displaying imply/max/min/normal deviation of a measurement taken over a number of oscilloscope traces. That is helpful if you want to see variation over a very long time.

The Math menu lets you outline new “digital” indicators as features of the digitized analog enter indicators. Along with the apparent ones (add, subtract, multiply, divide) you may compute squared, sq. roots, log, derivatives, integrals, high-pass-filter, and low-pass-filter primarily based on an enter sign. The integral and HPF and LPF options I discovered significantly helpful. Sadly there appears to be a bug within the HPF and LPF options, in that you just’re supposed to have the ability to modify the filter time fixed, but it surely appeared as if adjusting the filter time fixed did nothing in any respect. The integral and spinoff options ought to have had a user-adjustable fixed of integration/differentiation as properly; as a substitute, they seem to make use of successive show factors, which suggests while you change the horizontal scale to view over an extended timescale, the integral fixed modifications.

One characteristic that Hameg wants to repair on this scope is the flexibility to shortly modify a numeric setting with a really big selection. A few of the settings (for instance the HPF and LPF cutoffs) you sit there and spin and spin and spin the adjustment knob for a very long time earlier than you will get it to the place you need. Both they should put into place a nonlinear adjustment (when the adjustment knob is spun very quick, enhance the acquire of the adjustment change), or they want an information entry mechanism to enter constants digit by digit relatively than having to incrementally modify up or down.

Digital inputs

The connector on the entrance of the scope lets you join an elective HO3508 8-channel logic probe cable (about $400 listing value) to view digital indicators. That is fairly easy to make use of, and like different scopes with this characteristic, you may title particular person indicators so you are not restricted to only calling them D0, D1, D2, and many others.

I discovered two main flaws within the digital enter characteristic.

One is that, though you may modify the enter threshold of the digital indicators (TTL, 5V CMOS, ECL, and user-selectable ranges), you can not use separate thresholds for various digital inputs. This can be a very poor design alternative. Anybody who works on embedded methods lengthy sufficient will notice that you just usually discover each 3V and 5V digital indicators current, and this scope doesn’t provide a dependable approach to view these indicators concurrently digital inputs. It could even have been good to incorporate a selectable hysteresis stage (once more, on a per-input foundation) to deal with both noisy indicators or clear digital indicators for which it’s not straightforward to connect digital probes with out choosing up undesirable noise.

The opposite flaw is that as a way to view the digital indicators, you quit one of many analog channels. You possibly can both get analog channel 3 or the digital indicators, however not each. (With the higher-bandwidth scopes on this sequence that offer you 16 digital inputs, you lose one analog channel for every group of 8 digital inputs.) This can be a fairly uncooked deal. I do not know whether or not the analog indicators are at all times acquired, and it is simply that you may’t show channel 3 while you have a look at the digital inputs, or whether or not the scope stops buying channel 3 altogether while you activate the digital channels. Both means, it is a very poor design alternative.

One other characteristic that dissatisfied me was that the heart beat set off (set off on pulses with widths above or under or between limits) solely is relevant to the analog inputs, and never the digital inputs. Huh? This looks like an oversight that would have been prevented.

Serial buses

I regarded on the elective HOO10 serial decoder for UART/SPI/I2C indicators. This can be a pretty cheap choice at about $400 listing value. (For the Agilent MSOX3000 sequence, it prices $730 for UART and one other $730 for SPI and I2C decode.)

This selection lets you outline two bus indicators as features of the analog or digital inputs. So if you wish to decode UART, you inform the scope which sign to make use of, and if you wish to decode SPI, you inform the scope which sign is CLK, which is chip choose, and which is the information. I solely had an opportunity to check UART, and never SPI or I2C, but it surely’s a really beneficial characteristic to have. Within the previous days, after I labored on an SMBus undertaking, I can not keep in mind what number of occasions I’d spend 5-10 minutes poring over oscilloscope traces simply to strive to determine what information was being despatched, and I by no means knew whether or not I had made an error. As of late, with serial bus decode, you may show information values on the scope in a short time and save your beneficial time for engaged on extra vital issues.

The HOO10 lets you view information values (in binary, decimal, hex, or ASCII) and spark off particular information values. I discovered it very straightforward to spark off a single UART information worth; the scope had a characteristic to spark off a number of successive UART values, however I couldn’t get it to work, and I could not inform whether or not I wasn’t doing one thing proper or whether or not there was a bug within the set off perform.

Here is a screenshot of a UART decode, which was a really lengthy UART sign that I used to be in a position to set off and zoom to view particular person bytes.

After I first regarded over the spec sheet for the HOO10 I used to be apprehensive as a result of it solely confirmed the flexibility to decode one sign: UART RX or TX, however not each, and SPI MISO or MOSI, however not each. Nonetheless, that is per serial decode bus. You may have two buses you may decode and show, so can arrange one for UART RX and the opposite for UART TX, or one for SPI MISO and the opposite for SPI MOSI, or another mixture.

There may be additionally an elective module HOO12 for CAN/LIN decode for automotive functions. This retails for about $660, barely inexpensive than the MSOX3000 sequence CAN/LIN decode at $730. I used to be not in a position to check out this module.

Capturing information

Scopes are nice at viewing information, however usually you want to seize this information both as a picture or because the uncooked information. Within the Nineties we had been caught utilizing floppy disks or RS232 or GPIB to switch information to a PC. As of late scopes use USB thumb drives or LAN (Ethernet) community connections.

The HMO2024 has a built-in USB host port for plugging in a thumb drive, and comes with a USB machine connector and RS232 connector on the again, so you may hook up with a PC immediately. There are elective connectors for Ethernet (HO730: $317) and GPIB (HO740: $383). I attempted out the Ethernet connector, which is my most well-liked alternative, since fashionable scopes now have a builtin web-browser to make it straightforward to obtain information with out requiring any exterior software program.

This sequence of scopes does have a builtin internet browser for the Ethernet connection, and it is rather straightforward to obtain oscilloscope display pictures (just like the UART screenshot proven above). However there isn’t any characteristic for downloading the uncooked oscilloscope information by way of an internet browser. Hameg wants so as to add this characteristic – it is a must-have for my part, and I discovered it sorely missing. What you get as a substitute is a dialog field for coming into in SCPI instructions. I performed round with SCPI for some time and all I might handle is for the scope to report the date and time; not one of the different instructions I attempted labored.

After I gave up making an attempt to make use of the web-browser, I attempted out utilizing Hameg’s free oscilloscope communications software program, HMExplorer, which is meant to have the ability to obtain scope information as pictures or information in CSV format. I say “presupposed to” as a result of the software program is among the ten worst software program packages I’ve ever used. It has a bizarre consumer interface that could be very nonintuitive, and though I might get a CSV obtain began, the progress bar that’s displayed restarts many occasions (think about Porky Pig making an attempt to get out a sentence) earlier than it lastly manages to complete. I used to be in a position to obtain CSV information this fashion, however solely the smaller dataset (akin to the displayed information) and never the bigger “full” dataset; this system warned me that it would take a half hour to switch, and supplied no adjustable management for what number of samples to switch to the PC. I began on the lengthy switch, once more bought restart after restart, till lastly the information began to switch, after which it stalled at about 80%. After I tried to click on Cancel it instructed me that I must restart the instrument to renew communcations.

This can be a Home windows program, in order unhealthy as it’s, Mac and Linux customers are out of luck utterly.

The consumer’s handbook just isn’t significantly useful, and the interpretation from German to English is incomplete, with a number of German phrases scattered across the handbook and screenshots displaying this system with German-language dialog bins.

As an apart: what I do not perceive about semiconductor and check tools corporations from different nations that produce English documentation, is why they do not simply discover one or two vibrant faculty college students who’re native English audio system, and pay them $100 to proofread and proper their documentation, and take screenshots in English. Certain, actually good translation is dear, and I can sort of perceive poorly-translated English, but it surely should not value very a lot to get a local speaker to repair these manuals, and it will work wonders on firm status. After I learn a handbook with poorly-translated English, what sticks in my thoughts is that the corporate hasn’t bothered to make a high quality product. Merchandise from China are infamous for having unhealthy English documentation. I’d not have anticipated a European firm to run into this case.

So I gave up making an attempt to make use of HMExplorer.

This left me with the USB thumb drive as the one viable choice for saving uncooked waveform information. It was straightforward to get this information onto a thumb drive, and the time wanted to save lots of information onto a thumb drive could be very fast, solely a second or two. (One of many older Tektronix scopes, both the TPS2000 sequence or TDS3000 sequence, would take a particularly very long time to save lots of information onto USB thumb drive or CompactFlash card. I imply 15 or 20 seconds for perhaps 100K of knowledge. Boo.) My solely criticism with the thumb drive technique of saving information, is that the HMO2024 doesn’t provide a approach to save all the information into one CSV file. As a substitute you need to save one .CSV file per oscilloscope hint. This can be a poor design alternative that needs to be mounted. All of the analog information is being sampled concurrently, so there is no purpose the scope should not be capable of save one file with 1 column of time information and not less than the 4 columns of waveform information from the 4 enter channels. (if potential the digital inputs and math waveforms as properly.)

Miscellaneous

The one different factor that is neat about this scope, is that it comes with a pleasant assortment of check indicators. As a substitute of the same old sign and floor check factors for a 1kHz sq. wave, you get 5 check factors:

You possibly can choose within the scope whether or not you desire a 1kHz or 1MHz calibration waveform, or whether or not you desire a SPI/I2C/UART sign at varied bit charges, or whether or not you need digital outputs which can be counter bits or random patterns. That is very helpful to check out the oscilloscope performance. It bought me going shortly with the UART bus decoder, though it regarded just like the UART information was random; I’d have anticipated one thing predefined like a bunch of 0xAA55 information or a counter, or check information like sentences from Lewis Carroll’s “Jabberwocky”.

Not solely is the check sign characteristic helpful for making an attempt out this oscilloscope, however in case you are engaged on embedded methods and have to discover a supply of SPI/I2C/UART indicators, this scope provides that. (Maybe not as helpful because the arbitrary waveform generator choice out there on the Agilent MSOX3000 sequence, however you solely get 1 channel output from that scope, so SPI/I2C indicators are out of the query.)

The “Element Tester” characteristic regarded attention-grabbing (some sort of AC waveform utilized to find out part impedance by graphing present vs. voltage) however I did not have an opportunity to strive it out.

Different scopes on this sequence

Hameg sells two classes of oscilloscopes.

The HMO2024 is the highest finish of the lower-end (HMO2000) sequence, which provides selections of two or 4 channels and 70/100/150/200MHz bandwidth. (Only for reference: the listing value of the 2-channel 70MHz scope is roughly $1650, the 4-channel 70MHz scope is roughly $2050)

The upper-end HMO3000 sequence (HMO3032/3034/3042/3044/3052/3054) provides 2 or 4 channels and 300/400/500MHz bandwidth that’s “software-upgradable” just like the Agilent MSOX3000 sequence — the entire oscilloscopes are inherently 500MHz, however the bandwidth is factory-limited relying on how a lot you are prepared to pay. The principle variations between this sequence and the HMO2000 sequence apart from bandwidth appears to be that the HMO3000 sequence is barely bigger in depth and barely heavier, and the HMO3000 sequence provides 16 digital channels (in case you quit two analog channels) vs. the 8 digital channels on the HMO2000 sequence. In any other case the characteristic set seems to be about the identical.

The HMO3000 sequence is new and Hameg hasn’t printed a listing value.

Conclusions

The HMO2024 is an attention-grabbing low-cost oscilloscope with plenty of good options and a few room for enchancment.

Pluses:

  • Low value ($3000 for 4-analog-channel + 8 digital channel 200MHz)
  • Small footprint, light-weight
  • Good analog traits (low noise, hi-res mode, vertical offset past the display)
  • Intuitive UI
  • Zoom characteristic for exploring lengthy waveform acquisitions
  • Good decode/show of digital buses (UART/SPI/I2C/CAN/LIN) with cheap elective module
  • Straightforward obtain of screenshots with Ethernet internet browser
  • Straightforward switch of screenshots and CSV information with USB thumb drive
  • Devoted button for switching between regular and auto set off modes

Minuses:

  • Digital inputs not usable with out giving up 1 analog channel
  • Digital inputs not individually configurable with separate thresholds
  • No switch of waveform information by way of Ethernet internet browser
  • Very poor assist of USB/Ethernet information switch by way of HMExplorer
  • USB thumb drive switch of waveform information doesn’t present single-CSV file for all waveforms

At this level I am discovering it laborious to make up my thoughts. The Agilent MSOX3024 is a greater scope, however prices extra. The Hameg HMO2024 provides a value that nearly lets me afford 2 of them for the worth of the Agilent scope, and has higher analog sign dealing with than low-end Tektronix scopes, and a way more intuitive UI than Lecroy scopes — but it surely additionally has a number of disadvantages which can be sufficient of a stumbling block to make me hesitate earlier than wanting to buy it. 

My advice could be in case your finances is restrictive or you may dwell with the constraints I’ve talked about, the Hameg HMO2000 sequence is a very engaging worth, and it simply beats the Tektronix and Lecroy scopes I’ve tried. In case your finances is much less of a priority, and also you want the upper high quality, I would advocate the Agilent MSOX3000 sequence.

Blissful scoping!

P.S. There is a fairly good assessment of this oscilloscope on Youtube which covers a few of the options I have not, and you may see it in motion.

P.P.S. What is the worst software program I’ve ever used? It is the PC software program to make use of the Extech 380803 Energy Analyzer, which is a pleasant little briefcase-sized field that permits you to measure AC and DC energy flows. The final time I attempted utilizing this software program (2001? 2002?) it was atrocious 16-bit Home windows 3.1 software program, with poorly sized dialog bins, when the remainder of the world was utilizing Home windows 2000. One other merchandise in my high 10 is the software program for the now-discontinued Fluke Hydra 2600 sequence datalogger, which once more, was a pleasant piece of {hardware}, with kludgey software program that left a nasty style in my mouth. I do want check tools producers would view good auxiliary software program as an important a part of their accountability towards clients.

You may additionally like… (promoted content material)

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button