Open Source Electronics Lab for $30 | Hacker News Reader

Open Source Electronics Lab for $30

github.com

417 pointscolek4276 comments

This looks like a cool device for people just starting with electronics.

Slight OT: Its really sad to see that even though SDR really took off in the last couple of years, there really is no open source or at least low-cost toolchain for RF Engineering. No Signal Generation Hardware, no VNAs, no SAs... Mostly you have to make do with the still pretty expensive, 90s-Legacy devices of the large brands (Agilent/HP/Keysight or Rohde&Schwartz). Its also really stuff that Makerspaces and FabLabs are completely devoid of. Software is not much better, when you dont want to resort to unlicensed copies of the professional tools. Really hard to work on stuff like that when not either at a university or employed in that field...

I would have thought there would be a big central repository of Gnuradio models that cover all of the basics at this point--think Thingiverse for SDR--but I've never found it.

Maybe people are worried about the liability if some complete noob with a LimeSDR board starts jamming police radio bands or something?

When the LimeSDR Kickstarter was underway, one of the selling points was launching an app store based on Snaps on Ubuntu, I think this is the current version of that [1], which seems a bit empty to me.

GNURadio does also have a store directly for it called PyBOMBS [2] as well, though it is unfortunately not included by default, and it has been a few years since I checked it so I don't know how much it has.

Gnuradio is a bit overcomplicated at this point also, using both Qt and GTK, and a long list of other dependencies.

Luaradio[2] is quite nice as an alternative, and is much easier to embed into an end application as Lua itself is very good at that kind of thing.

[1] https://limenet.snapcraft.io/ [2] hhttps://github.com/gnuradio/pybombs [3] http://luaradio.io/

No. 2 takes a while to setup IIRC. It isn't really click to run. Though once it's setup it's pretty straightforward to use. AppImage, snapcraft etc. would help a lot here.

That was the same thoughts I had on gnuradio + a cheap sdr.

I remember I tried to get a temperature sensor decoding gnu radio plugin working and I never solved it.

Couldn't run it on a Pi, it required a fast intel system to be usable. That eliminated the nice raspberry pi "canned image" solution other projects use to cut through the dependency mess.

So I had to do a linux installation from scratch.

But the online packages for gnu radio were broken or out of date.

So I had to build GNU radio from scratch.

And once I got GNU radio going any existing gnu radio program was broken because it required an exact version of gnu radio and other dependencies.

Maybe it is better now, that was a few years back.

I've also come to see the light with Arch Linux, so that might help.

On the plus side, those legacy devices are orders of magnitude better than anything that a beginner could build and can be bought relatively cheap through surplus stores.

Not sure if anyone in here subscribes to The Signal Path [1] on YouTube, but he does tons of teardowns (and repairs!) of RF/signal processing hardware new and old. You'll see precisely what jaquesm is talking about. The marquee specs might not look so great on paper, but you'll find absurd levels of craft and attention to detail in some of that older equipment.

[1] - https://www.youtube.com/user/TheSignalPathBlog

Love watching the signal path, especially the pooch appearances :)

High dynamic range (i.e. low noise figure, high linearity, and especially shielding) and low phase noise will always be expensive. They are independent of software and Moore’s law.

The design software is not cheap either, as typical software skills are only 1/3 the knowledge required to develop that stuff. Also need circuit theory, EM, scientific computing, etc. There will probably never be an open source equivalent.

I have an RF/Microwave lab at home with HP VNA, spectrum analyzer, power meters, signal generators, and Tek scopes and curve tracer. Also Microwave Office and Altium.

Yes, that stuff is expensive, but has also never been cheaper to buy. An 8566B spectrum analyzer can be had for $2k, and is far better than shit from Rigol. That 8566B cost $85k in 80s/90s dollars.

There will probably never be an open source equivalent.

That may be a little pessimistic. QUCS development seems to have stalled, but they were certainly on the right track.

Given the kind of sponsorship that, say, KiCad enjoys from CERN, I would expect some pretty awesome RF engineering tools to emerge.

QUCS is nowhere near equivalent to Microwave Office or ADS; those each have well over 100k man-hours of code development. Harmonic balance, transient for distributed circuits, envelope and system simulation, layout, 2D and 3D EM, and hundreds of models, all tightly integrated. Not to mention all the MMIC and RFIC PDKs.

It’s fine for hobby use, but if you need to make $ and get designs out the door, you need to use the real stuff.

I tried KiCad, but using Altium for 25 years, it just doesn’t have the functionality I need. Lacks basic things such as assigning nets to arbitrary polygons, which is a must for RF.

Is QUCS the current best open source equivalent? I used to do hardware long time ago and might play around again as a side project.

redpitaya [1] makes an nice modern SDR transceiver. [1] https://www.redpitaya.com/

Higher barriers for entry may be good for SDR. If it were too easy there may be problems with inexperienced or inexperienced+nefarious people causing problems. Spectrum is a limited resources and it is fairly easy to poison the well.

If that happened too much regulators might tighten up and make it impossible to do (or sell) any SDR.

The barriers are exceptionally high though. I have a degree in Physics as well as a degree in Electrical Engineering with a specialization in RF from a well-regarded engineering program and wasn't able to find any jobs in the field. In SW engineering, I am able to get offers for high paying jobs inside of a month.

I feel like I would be able to get an RF job much more easily if I could build a portfolio like I did for SW. It's a moot point by now probably; I'm already getting paid more than the average general practitioner while being able to work from home 100% and wouldn't want to revert to lower pay while being stuck in a cubicle all day.

The ARRL has a quarterly publication, QEX, that frequently publishes FOSS hw/sw designs for VNA, spectrum analyzers, and frequency synthesizers.

http://www.arrl.org/this-month-in-qex

American Radio Relay League (ARRL)

Very complicated web site that a defense contractor would love. But definitely worth looking into. Wow.

I vaguely recall this group as serious ham radio people who know how to keep communication going even when disaster might take down other networks. We have other technical volunteer groups out here, like Search and Rescue, first aid and so on.

What about LimeSDR?

You can get by with an oscilloscope and function generator both of which have gotten a lot better. A lot of homebrewing aims at using those devices. This is in the HD worls.

For the scope part:

- This board is 750kSPS, the most well known cheap alternative (not necessarily open source), is something from JYeTech, e.g., DSO-138 [1] (it's 1MSPS)

- Both 750kSPS and 1MSPS are hopelessly inadequate for serious work

- In the USB scope arena, the two top contenders are OWON VDS 1022i, or Hantek 6022 variant (I prefer 1022i; it's 100MSPS [2], I think Hantek is 48MSPS).

- Finally, the best bang for the buck, and open source, is the Haasoscope v9 selling on ebay (directly from the creator, Andy Haas, who's a physicist and wanted to create a cheap but good open source scope). The bundle is $139 [3], but there is an add-on that further increase the frame-rate or something that's $40 or something, though I consider that to be essential not an add-on. (it's 250MSPS)

- if you wanna get a new and stand-alone scope, Rigol ds1054z $375 [4] is a hot favorite (also watch out for Keysight, they're trying to get into low-cost scope market). 1GSPS.

Disclaimier: No affiliation.

[1] https://jyetech.com/dso138mini-diy-kit/

[2] https://www.amazon.com/dp/B00HC4KY2G

[3] https://www.ebay.com/itm/283330921663

[4] https://www.amazon.com/dp/B012938E76

I have a rigol MSO1104 (1 ghz sampling, 100 mhz analog bandwidth, logic analyzer) I bought after I retired for doing some work out of my house. I remember it being about $1000. It's very nice, I would highly recommend it for a home lab. I also bought an analog discovery 2 to bring to the library where I teach Arduino and Raspberry Pi classes. It's 2 channel, I can't remember the bandwidth, I think 60 mhz analog bandwidth and maybe 500 megasamples/second? Anyway I think I bought the pro bundle, a little less than $300. I think it's a very nice setup for someone to get started, it has lots of really cool features like a network analyzer, minijack input for audio, serial decoders etc. The PC software works good, available on linux. It's small runs off USB power. I know the probes don't have the isolation that a normal commercial scope does, but for a lot of hobby/learning type things it doesn't matter. I like the idea of making your own hardware and learning things from it, but for a beginner in electronics or the occasional user I think the analog discovery is a pretty good value. I was never a big fan of National Instruments when I was working, but I think this is a pretty good product. https://store.digilentinc.com/analog-discovery-2-pro-bundle/

(edit) - actually it's 30mhz with the bnc probes.

A two channel logic analyzer seems very inadequate. Need four minimum.

> Both 750kSPS and 1MSPS are hopelessly inadequate for serious work

It's perfectly fine if the serious work you're doing is audio.

Appreciate the detail.

I'll add that I can recommend the DSO-138 for those new to electronics and want to start with something fun. It comes in a kit (soldering + assembly required), and is great way to get introduced to electronics assembly. At the end you'll hopefully get a working oscilloscope to play with :)

It's an awesome project for kids too!

Speaking of Hantek, has anyone tried their handheld oscilloscope/function generator?

https://www.banggood.com/Hantek-3in1-Digital-OscilloscopeWav...

I am a fan of these types of small multifunctional electronics equipments. I think every hobbyist should have one or more of these to play with. Here are some of them;

ExpEYES - Abbreviation for "Experiments for Young Engineers and Scientists" (http://expeyes.in/) from IUAC, New Delhi, India. It is quite affordable.

Redpitaya - https://www.redpitaya.com/ Can also be used to learn FPGA programming.

Analog Discovery 2 & Digital Discovery - https://store.digilentinc.com/usb-scopes-instruments-circuit...

ADALM2000 - https://www.analog.com/en/design-center/evaluation-hardware-... Can also be used to learn FPGA programming.

This looks fantastic for quick and dirty audio work, especially hobbyist pedal building/prototyping.

A teeny feature request for the multimeter would be an easy way to make a spreadsheet out of multiple measurements, and/or transistor parameter measurements. Since the bandwidth is so low, I'd imagine a lot of users will use the scope to hand match transistors.

Agreed. I started tinkering in this space over the last year (done loads of software, started trying to make my own synth). The lack of an oscilloscope basically makes it impossible to get anything done - you're kinda just working on hunches.

People kept recommending the RIGOL DS1054Z, but at £300 or so it's too punchy to invest in unless you know you're really going to get a chance to use it. Will definitely try one of these little units.

That's like doing engine work blindfolded!

Amateur tip though, lots of good deals on used scopes out there. Companies seem to cycle through them, probably a calibration thing. I've seen 200MHz scopes go for < $150 USD on Ebay in the past, (inconveniently enough at times when I didn't have $150 to blow for an older Tektronix).

For audio and most 8 bit work you can buy a two channel 10Mhz picoscope for about £120. Really solid piece of kit, and it's designed in the UK (if that has any weight). Has tons of stuff like signal decoding built into the UI which you have to pay for (or hack) on the Rigol. Don't believe the crap about USB scopes being bad. It's really handy for not taking up valuable bench space, and you can take it in the field easily. It also has a basic signal gen, but I've not used it much.

I also have a 1054 which is great - sometimes you just need 4 channels. I would be happy with either.

The trade off is something you can fit in your pocket versus two extra channels and more bandwidth. The 4 channel 50Mhz version is actually quite close to the Rigol price.

https://www.picotech.com/oscilloscope/2000/picoscope-2000-sp...

I have the 2204A and can second the recommendation. Really good bang for the buck (120EUR here in Germany, including two probes), very compact and robust. Of course, with 10MHz you shouldn't expect too much, but for looking at low-speed communication it is perfectly OK. What I really enjoy are the many different trigger options (and they actually work). When buying an USB scope one must remember that it is nothing without the software, and the one from Picotech is well done. They also support Linux although that one is still in Beta and still has less features, but they say they are working on feature parity. It's not (F)OSS, though.

Look into the ADI Analog Discovery 2 for audio work, as it has a Bode analyzer and THD in the spectrum analyzer.

https://youtu.be/1rOvdsxQVTY

Something I would add to this is the functionality of a component tester.

If you don’t know what they are for, they are cheap little arduino based devices that you can plug a lot of basic components into, and have them identify the component for you. Plug in a resistor, see the resistance. Also works for mystery transistors as well as caps and inductors plus diodes and other things.

They are fantastic if you’re one of those people who has a tendency to collect a lot of components but forget what they are.

Here’s an amazon link to a typical one...

https://www.amazon.com/dp/B071Y5CHPK

Here is the original source for the project: https://www.mikrocontroller.net/articles/AVR_Transistorteste...

It is not really based on Arduino, but it indeed is pretty nifty little gadget. The documentation goes into great detail on the operating principle, how it detects and measures different components.

Ah thank you. I thought they had roots in some open source project.

This thing has been pretty key to a lot of my hobby electronics projects.

I was hoping this was an open source alternative for the typical analog or mixed electronics benches.

As a physics student I was quite dissatisfied with these benches for multiple reasons:

1. They are far from educational, the course notes and lab notes for experimentation classes are much better at demonstrating principles. The books that come with the analog bench "toys" lack the ambition to actually teach, and avoid formulae, at least the books could contain or link to an alternative track of circuits which does have the ambition to teach electronics.

2. A lot of the circuits simply don't work, I believe this is due to occasionally substituting a part with a cheaper part with slightly different specifications, such that after a while of inevitable sales to parents who don't give feedback, the supplied components don't work in the proposed circuit, the original designer of the circuit would notice that the current parts can't produce the intended effect (chirping birds that give popping sounds instead of chirping sounds etc).

3. Most of the analog electronics benches use springs as electrical nodes, I believe today stackable jumper wires would be a better choice.

I think if enough electrical engineers with kids band together they could easily design an open source analog / mixed electronic bench, and properly document the required properties for the parts, and design it as a PCB with headers for jumper wires...

I was in the process of buying the board when I got the Database Error. Came to check Hacker News and discovered why!

This is something I need. What alternatives exist? 30 dollars isn't bad, but communities of support are also nice.

The Red Pitaya STEMlab is a considerably more expensive device, but it's more than just a toy. It has two 14-bit, 125MS/s ADCs and DACs, 16 digital GPIOs and gigabit ethernet.

https://www.redpitaya.com/

That's nice, but I imagine the use cases are pretty different, the Pitaya's sampling frequency has 3 orders of magnitude on the Labrador (and 2 more bits of sensitivity). I think they're both pretty impressive for their respective prices.

Why is the LCR meter expansion so damn expensive? The base starter kit is pretty reasonably priced, but I can't figure out what is going on with that LCR meter thing.. based on pictures the HW doesn't look like it has anything special on-board, so is it just a SW license cost thing?

The problem with RedPitaya is extremely poor software support. And without that you have only an expensive brick.

Worse, the board itself isn't open source, AFAIK, so even stuff like finding a schematic to get things done could be a challenge.

Someone recently mentioned something called “LHT00SU1”. It’s primarily a logic analyzer, but it has one analog input. (I’m not into EE, never used scopes, just randomly found a discussion about it recently)

In that vein the Saleae Logic[1] are really awesome, the prices went up recently but they're the real deal, next step up from them gets into the $X,000/$XX,000 range very quickly.

[1] https://www.saleae.com/

I have one of the Saleae devices, and it is wonderful. I still like (and sometimes really need) my more traditional full-size equipment, but the Saleae was inexpensive and it's easy to whip out and hook up to something.

I don't know if they still have a hobbyist discount, but when I bought mine they knocked a good chunk off the price after I emailed them and told them my plans for it as a tinkerer.

Wow they've really jacked the price up!

No disrespect to Saleae's Logic as even their 1st gen product was indeed outstanding value (and may still be at the current price point; software front end is well executed). But to be sure, there's still a significant hardware gap that sets it apart from "the real deal".

I got several working, albeit, old, oscopes and signal generators from pawn shops for $50 to $100 each. Had to search a while, but they tend not to sell well there, so they eventually get marked down.

I don't know about how much community support they have, but seeed sells some alternatives:

https://www.seeedstudio.com/oscilloscopes-c-880/DSCope-p-242... https://www.seeedstudio.com/oscilloscopes-c-880/Xminilab-Por...

Well, the next step above would maybe be a low-end PicoScope (they start at 115$ for 10MHz).

https://www.picotech.com/products/oscilloscope

Try http://expeyes.in/ for a better alternative. Take a look at my other post in this thread for some more costlier options.

But if you actually need a scope you can buy one in a case with a rechargeable battery, display, and 20MS/s, 5 MHz bandwidth, from Banggood for USD 40. https://www.banggood.com/DANIU-ADS2050H-Handheld-Oscilloscop...

No idea how good it is though.

I don't mean to disparage the Labrador but I suspect it will be more use to people who integrate it into some other project than to those who need a scope.

The "All Sun EM125" Multimeter+Oscilloscope (https://www.aliexpress.com/wholesale?catId=0&initiative_id=S...) is far better for a little bit more money. The specs are very good for the price. On the minus side, the oscilloscope doesn't have storage and the display is monochrome low resolution. However i have found it very convenient and useful in actual usage.

"All Sun" also has a whole range of handheld multi-functional scopes which seem pretty good.

Website[0] is down?

> Database Error

> Error establishing a database connection

[0] https://espotek.com/labrador/

Here is the internet archive link: https://web.archive.org/web/20180324194926/http://espotek.co...

Review on YouTube: https://youtu.be/7xwFEvJxv40

This came out in 2016. Is it gone now?

Reasonable idea. More protection circuitry to prevent blowing out the connected PC might be a good idea.

optical isolation, look for highspeed optocouplers for isolation, and something like digital potentiometers are good too for clamping an abitrary analog voltage using a digital mechanism.

In the spirit of the discussion of low cost lab electronics - may I throw in the Analog discovery 2[0]:

[0] https://analogdiscovery.com/

"low cost" == $279? Sure, if you're a professional.

Well, 100x the bandwidth for 10x the price.

They have discounts for educational institutions if price is really that much of a concern, it's great value for the features they pack in that little thing...and honestly "professionals" use much much more expensive instruments in their labs.

Others, from Analog Devices:

ADALM 1000 [0] (~$47 from Mouser)

ADALM 2000 [1] ($99 student pricing, from Analog)

ADALM Pluto (SDR) [2] ($136 from Digikey)

The ADALM 1000 is an evolution of the Nonolith Labs CEE [3] (after it was acquired by Analog Design).

The ADALM 2000 is an evolution of the ADALM 1000, both ADALM 2000 and ADALM Pluto use a Xilinx Zynq (FPGA + ARM core).

Have no connection with the company, other than being a happy user.

[0] https://www.analog.com/en/design-center/evaluation-hardware-... [1] https://www.analog.com/en/design-center/evaluation-hardware-... [2] https://www.analog.com/en/design-center/evaluation-hardware-... [3] http://wiki.nonolithlabs.com/cee/

Is the ADALM1000 really worth it? Seems too basic and limited to me. I have had my eye on the ADALM2000/ADALM PLUTO for a while now (already own a Digilent AD2 and Redpitaya) but haven't committed to buying them.

It's a pity the software is distributed in binary form only, which in this case means no RPi support.

This seems too good to be true. Is there any reason an embedded noob would not find this did what they needed for small projects?

I guess you still need a breadboard/uC/programmer/wires, but still

Edit: free shipping? Shipping to Canada might cost as much as the device. The price is too low if it’s actually worth using

As one who is slightly past the noob stage in embedded systems, my impression is that the device can't run a stored program on board. In other words, it's not an Arduino.

At the same time, an Arduino is not this. An Arduino can make measurements and generate signals, but doesn't come with pre-written support software if you just want to use it for that purpose.

For this reason, you might actually use something like this electronics lab alongside something like an Arduino, and use one to observe and measure the behavior of the other.

Even for a "embedded noob", the specs are not that useful. You need something better. See my other posts in this thread for alternatives.

Does it have both ESR probe and a spectrum analyzer? I just recently bought a RTL-SDR v3 USB software defined radio for 30 bucks and it seems to function as both, in addition to being pretty kick ass SDR for the price.

I have been following a promising project[0] for quite long. They haven't yet released the product though.

[0] https://madresistor.com/box0/

As a total noob what can I do with this and where should I start

It gives the following very basic pieces of equipment for doing hardware development. However this is quite low end and will have lots of limitations. A high end logic analyzer could be $50k for example.

USB Oscilloscope - observe an analog output waveform

Signal Generator - supply an input waveform

Power Supply - power the device

Logic Analyzer - observe digital signals

Multimeter - get average voltages

This is going to sound really basic, but a good start might be just to reproducibly generate and measure some signals: DC, AC, etc. Make an LED blink. Measure the voltage across the LED as a function of time. Get used to seeing a graph of voltage versus time, and interpreting it. Hook it up to a microphone and talk to it. Measure the headphone output of your PC.

This is all about just learning to think in terms of signals (static and as a function of time) and how they are measured. Try to predict what's going to happen, and see what you get.

You learn hands-on Electronics (and Embedded Systems) using these types of equipment. If you can afford it i would suggest Digilent's "Analog Discovery 2" (https://store.digilentinc.com/usb-scopes-instruments-circuit...) Digilent has lots of online course material (https://reference.digilentinc.com/reference/instrumentation/...) which you can follow-along to learn Electronics.

Sorry to nag. For a project that I am working on:

Does any of this boards have an oscilloscope with four channels?

The haasoscope mentioned earlier does https://www.ebay.com/itm/Haasoscope-open-source-4-channel-ex...

Reply on news.ycombinator.com