Slashdot Mirror
Ask Slashdot: What Equipment and Furniture For an Electronics Hardware Lab?
bartoku writes "Slashdot, what would you put in your dream electronics hardware lab? I am putting one together, and I'm looking for suggestions on everything from equipment to furniture. My aim is for a professional-grade setup, not just a hobby lab. The goal is to be able to test and debug modern electronic device prototypes. I would love to see money-is-no-objective suggestions alongside more economically practical solutions. Links or contacts for good distributors to acquire the equipment and furniture are also welcome. I'm also interested in commentary on renting versus buying new or used higher-end equipment to be economical and keep up with equipment that will become obsolete quickly."
Check out David Jones' EEVblog, particularly episode 168. http://www.eevblog.com/
Conductive flooring paired with electrostatic discharge heelstraps (or better yet static dissipative shoes) will go a long way toward mitigating ESD risks in your lab. While wrist straps are effective they are inconvenient and therefore more likely not to be used consistently. Most lab furniture is conductive, but you often pay a bit extra for chairs with conductive castors. The need for lab coats depends on the apparel your lab staff typically wear (wool and synthetics should be covered with a dissipative lab coat, cottons are not typically an ESD risk).
My opinion? See above.
1 is ok, but a Tesla coil or Jacob's Ladder would be a better bet.
Not a sentence!
Exactly. This question seems just like a poseur question "what tools do I need to build an F1 car in my garage" or "what tools do I need to build lots of great furniture".
As for "furniture" - get a decent bench and some shelves. What do you want, padded 60's lounge chairs to feel like a mad scientist? And then once you have said shelves and bench, once you actually find the need to DO something and don't have the tool to do it: then buy it. Stocking a home lab full of shit you'll never use is complete rich nerd masturbation exercise.
By no means is my workshop the coolest in the world, but its a combination of years of experience, building and designing, and this is how it is:
1) Raaco shelves, these are absolutely essential, youd hate to run out of components in the middle of a project, so you need these, fill the walls! http://images.toolstop.co.uk/product/6651eea4432e327d9f2017ea860bef09.jpg
2) You need HEAPS of components. Now, youre probably not a millionaire, if you where...you wouldnt ask us geeks, youd just purchase whatever, so here is how I get my stash. I go to ham-fests, the radio amateurs usually have thousands if not millions of surplus components theyve grabbed from a run-down electronics shop or factory closedowns. Make a HUGE list of your essentials, and go collecting. Itll take a few years, but youll get there. I have MILLIONS and MILLIONS of NOS (new old stock) components from all over the world by now. Ebay is your friend, but beware of FAKE components, expensive components sold for peanuts...could be fakes, but its still relatively rare imho. Go hunting for closedowns of electronics labs, stores and much more, 70% of my components comes from there, and usually for pocket-change. Hang out...befriend the managers...listen and pay attention. Before you know it, youre the "buddy" who gets everything for nothing.
3) Get SMD reels too. Have a copy of your DIL/DIP discretes as SMD equivalents, this is when youre finished prototyping with the discretes. You need the full size discretes in order to experiment properly. Far too many wannabe designers design everything in CAD and scratch their heads endlessly over their designs, lacking on-hands experience with the easy to handle components. This is understated today. A lab like this is essential for quick and good development.
4) You need ROCK SOLID tables rather than fancy glass tables, so purchase some old super-solid office equipment rather than shop IKEA. Sometimes youll throw a 50-100 kgs of instrumentation on your table, and bye bye IKEA. And itll get dirty, and itll drown in solder waste (which you will eventually get everywhere). So it must be a surface solid and easy to clean.
5) You wall should also have a tool-rack, here you need the rough tools such as screwdrivers, mini drills, bits, cutters, pliers and whatnot. Youll also need some hangers for your endless numerous test-cables. Hang the test cables within easy reach so you can keep your shop tidy and neat. This will become more important than you may think.
6) Speaking of which, numerous of testcables you need (Yoda talk)... banana plugs, soft-silicone cables for power connections, extendable banana plug cables are essential, dont skimp on quality here. In fact, you may nearly skimp on everything except this. Test cables are notorious for going bust, and killing that spirit when you finally discover that you bought cheap crap...and spent hours just to find out your test cable is leaky, crappy and such. You need 100mhz range probes, probably higher...and more expensive, but start out with common 100-250mhz scope probes.
7) As for test instruments, you need these basic things: 2 Benchtop multimeters, 1 portable multimeter, 1 frequency counter (min 2.6 ghz), 2 Benchtop oscilloscopes, preferably one analog and one digital...Ive got 4 of them for various reasons...you can never get enough scopes and multimeters. Function generator is essential for repair and design, a 10mhz will do, preferably with TTL level output as well as variable analog. Get a Signal generator too, 1 ghz minimum...the 1+ghz something...needs to be very stable if you operate above these frequencies. Benchtop lab PSUs... get some with both analog and digital readouts, the older generation analog psus tend to be less noisy and better at delivering at high peaks. Switchmode PSUs are needed for those higher power needs, but have at least one of each.
It doesnt matter what its for... Well it does, but a high bench and stools rather than desks and chairs will be highly beneficial. It is far more comfortable to work for long periods when you don't have to bend down all day. I always preferred a single flat island surface (no joins) in the middle of the lab with plenty of power. Bigger is better.
Plenty of racking or shelving for storing equipment and drawers or similar for smaller things.
Oh, and a shitload of light. When you think you have enough light, double it. You can always remove half the globes/tubes, but if you don't have enluh fixtures you will be constantly working in your own shadow and that sucks.
And every lab needs a machine that goes ping. Why it goes ping is, of course, dependent on the use of your lab.
I cannot stress lighting enough.
Almost everyone gets it wrong, and it's the most annoying to fix after the fact. You want whatever room you're in bright - very bright, and illuminated uniformly. If you're putting in shelves or anything that occludes light, plan to have lights attached to it. You'll find yourself *much* happier later.
The flippant answer to your question is that you should get whatever equipment you think will be useful for whatever projects you're doing. If you don't know what you want, then you won't be putting it to use anyway. "Electronics" is not one discipline. It's a collection of related but different fields, like different specialties in medicine. What equipment does an operating room need? Aside from a few basics, the answer depends entirely on what kind of surgery is being performed. Having a network analyzer or a service monitor in your electronics lab is great for some types of work, but if you don't already think you'll be needing those then they're just going to collect dust anyway.
I do a variety of different kinds of electronics work, but most of it is RF (ham radio), high voltage (Tesla coils, fun plasma experiments), or high power (switching power supplies), or all three (induction heating, BIG lasers, serious radio transmitters, kick-ass solid state Tesla coils, etc.). This requires an array of tools and equipment that ranges from common and universal to highly specialized. Here are my key assets:
- Fluke Scopemeter 199C, 200MHz portable digital oscilloscope. If I could keep only one test instrument, this is it. Totally worth the $4k. I literally could not do much of what I do without this tool or something similar. I love my Scopemeter. It's just the right combination of portability, durability, and signal analysis capability. The electrical isolation of the fully-insulated battery-operated unit is a huge benefit sometimes too.
- Fluke 287 digital multimeter. A very high-functioning DMM for general purpose use. RMS readings on funky waveforms over a wide frequency range. Accurate measurement of component values.
- Klein CL2000 clamp-on AC/DC ammeter and multimeter. A really great tool for general DMM use, as well as non-contact RMS measurement of high currents, both AC and DC. This bad boy can accurately measure how much current your car draws while cranking the starter, or the true RMS current of an arc welder.
- RigExpert AA-520 antenna analyzer. A rudimentary but powerful digital antenna analyzer for HF through UHF frequencies. Great for its basic purpose, but also capable of doing lots of neat tricks like tuning duplexers if you get creative with it.
- Cheap pocket DMM's. They're like $20 and it's great to have several sitting around to use as monitors for various parameters on a system during testing. You can blow them up or fry them with an RF field and not feel too bad, saving your precious Fluke gear.
- Solomon temp-controlled soldering station. Needs no explanation. I feel that there's no need to go crazy on soldering equipment. A sub-$100 station will do the job just fine, even for fine SMD work as long as it's of decent quality. PID temperature control, low mass, and a hefty heater are all requirements.
- Granite work surface. Of all the surfaces I've worked on, I have found granite to be the best. Preferably pure black so things show up on it. It is heatproof, electrically insulating enough for any purpose, anti-static, strong, hard, and pretty affordable compared to a digital oscilloscope. A couple hundred bucks will get you a very good slab section to work on.
- Automated external defibrillator. I often work on circuits that can kill me with one false move. Having an AED nearby and showing friends and family how to use it could save my life some day. Very rapid defibrillation (within 5 minutes, preferably 2) is the ONLY effective life-saving treatment for electrocution. With very fast intervention, the chance of survival is excellent. By the time an ambulance arrives it is far too late. It's $800 well spent if it even gives me a 1% chance of not dying. Skip this if your work doesn't involve much line-voltage or higher.
I am a geek attorney, but not your geek attorney unless you've already retained me. This is not legal advice.
A opposed to fake earth grounded mains?
Yes exactly. A "3-hole outlet" will work perfectly well at powering things if you don't hook up the ground. Its ungrounded, but you won't find out until something shorts to a chassis and you get electrocuted.
Another prime failure mode for older houses is relying on conduit for ground path and having j6p previous loanowner replace or remove a chunk of the conduit leading. No more ground anymore.
One funny failure mode of conduit grounds is not being able to source/sink 15 amps to blow the circuit breaker. Been there seen that. So hot wire shorts to chassis, resistance of the ground path is so immense from poor/corroded connections that it only drops 10 amps or so, until the fire starts or someone gets electrocuted anyway.
Finally IF you're doing analog or analog-ish stuff and you think grounding will cut down on noise, a "bad ground" might act as an antenna and make it even worse.
You can buy a little plug in doohickey from home depot or whatever that costs like $5 full of neon bulbs that will tell you if an outlet has power, if the outlet has hot-neutral reversed, and if theres a ground. You could build one in about 10 minutes using a little project box and a fat stack of 120 volt lamp/indicators/neons/whatever. The point being that you better not have 120VAC between neutral and ground, etc.
I would estimate from experience that given "normal lifestyle" with J6P fooling around with his own house wiring that about 5% of outlets will get mis-wired per decade, unless J6P had an electrician buddy or was smart enough to drop $5 on the little outlet tester lamp thingy.
So now that I've explained how having a 3-prong outlet doesn't mean you have a ground at all, we can move on to explaining exactly how, why, and when you need an isolation transformer on your test bench. The answer is short, if you can't explain, in detail, exactly whats going on WRT ground loops or floating gear to do HV work, you are not supposed to be using that stuff (TLDR is sometimes "grounded" test equipment needs to be "ungrounded", although its somewhat risky sometimes its the only way).
"Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
A real ground is nice, but if you are working with an ESD strap, please make sure to have all the bench outlets on GFCI!!
Don't do anything really really stupid like driving a ground rod just for your lab and tying your ground pin to that; make sure it is bonded to the main ground.
Yes and no. In the case of digital logic you can write a model in VHDL, Verilog or whatever floats your boat. And then do functional verification on that. Dump it in a FPGA or hope your computer is fast enough. But for analog circuits it's not as easy. SPICE can give you a pretty good idea, but it's not perfect. Even more advanced programs like ADS won't give you the full picture. As the final step you will always need to actually make a prototype and just hope it works.
So yes you can test small parts of the design, but a large design can't be simulated without significant computing resources.