I am building brand new, 5K$ lab for a small startup, Judge my choices

[Placeholder]

I’ve always been a hobbyist, tinkering with electronics for about 8 years by now, (I am 24). Also, I worked for non specified German engineering company as a “Associated engineer” for 2 years. Nevertheless, I am today building a new lab, for a startup I co-founded. And because I lost most of my equipment (details how or why are irrelevant for this post), I have an unique opportunity to get “tabula rasa” lab. To build it from scratch. I will list what I think of each instrument and why I decided to go with it.

The purpose of the lab — Our company is 95% software, but there is going to be need for a custom HW product to be developed, I don’t need to be afraid to be specific — Something that can host an NVIDIA Jetson SOM, maybe few connectors, powersupply is trough USB C, nothing too complex to develop, but certainly not easy to validate and prepare for certifications.

Budget is 5-7.5K$ that will expand further after we get trough prototype manufacturing. Nevertheless, this is not supposted to be completely standalone lab, I do have access to advanced measurement equipment in the a testing center, (paid hourly of course) if I need to verify some advanced highspeed




Soldering iron(s)
—————————
175$ TS1C wireless soldering iron — for daily soldering tasks, with added convenience
26$ Pinecil — for heavy duty soldering or when it is necessary to use two soldering irons at once
~20$ spare tips for both
(Why?) I have tried it in person, and the convenience to just pick up a soldering iron and run around desolder some wire on a big prototype beyond the table is great. I did work with JBC most of the time when I was working in a corporation, they are the best, period. but I can’t justify the added value in my workshop.

200$ YIHUA 853aaa — cheap rework station

((fume extractor… Just a fan with a aluminium pipe right outside the office window, too expensive to filter, and the cheap aren’t worth it))

Soldering consumables
—————————
18$ TF-V3-XXX-K Flux — apparently best flux known to man, yes I will buy from digikey and not ebay
11$ SMD291 ChipQuik Flux — also great, but cheaper, I want to have something to compare to.
(Might choose one or the other if audience suggests their favourite, and not buy both)

20$ superWick from MG Chemicals — more of an experiment, my friend swears by this strange piece of “luxury consumable” I want to try and have something to compare
1$ wick Aliexpress grade  — of course…

~50$ SN100C AIM lead-free solder — worked with this personally
~50$ K100LD Kester lead-free solder — my friend preferst to SN100C
(Deciding between those two)
(Please do not suggest leaded solder, let that be history, I worked with leaded and unleaded, and yes, it is slightly more inconvenient, and the joints are more brittle, but I am not designing MIL grade, I don’t need to have a health hazard on the table)

30$ SMDLTLFP10T5 low melt point solder paste

color wires from adafruit (I already have)
Enameled Copper Wire adafruit (I already have)


general equipment
—————————
30$ Hakko Depaneling tool — Might be bit too much of an luxury?
(Why?) Given the low quantity we want to manufacture, and that they are exclusively made for the Czech market, It makes sense to test, depanel, and assemble the products here in our offices, and this might speed up the process slightly

wire strippers (I already have)
wire cutters (I already have)
Utility cutter knife (I already have)
Tweezers (I already have) — ceramic tip tweezers to be specific, multiple pieces
Capton tape (I already have)

100$ sensepeek 6012 — mainly as PCB holder, with with the added multimeter probes for some very quick debuging
(Why) I am not buying the oscilloscope probes? I feel like they are bit too overpriced, and altrough I had this system at work, you can’t measure from both sides, making it somewhat redundant

20$ generic usb microscope from Aliexpress

10$ SMD mat + straps

((This might be a blind spot, does anyone want to suggest any other small hand tools they couldnt exist without?))



Power supply & Load
—————————
640$ GPP-2323 — (or 3323) Programmable Power supply & Load
(Why?) I need to test load more than just occasionally, having a dual power supply that can also be switched into dual load, or single source, single load, is amazing
((What I don’t like is the resest connectors, the reverse engineering I found here is solid to me for that pricel))

((I still feel I am being too cheap here, is competition from Siglent or Rigol better?, even if it means having two machines, source, and load? Am I making the right choice with GWInstek?))

30$ banana cables, multiple lenghts



Oscilloscope
—————————

I tested IN PERSON, MSO5074, SDS2104X and InfiniiVision1000X
I absolutely HATE the MSO, the user interface is so bad it would be in severe danger of being thrown out of a window.
InfiniiVision1000X is just ridicolously expensive ((and unhackable))

1200$ SDS2104X (with maybe slight fw modifications) — All the features I need for 95% of my measurements, good sample rate, good BW, signal generator

((Does anyone have a contender against the SDS2104X? Does anyone feel that I should go much higher in budget for osc?))

~10$ Bare BNC probes — I find probing with manual probes too unreliable, usually I just solder a coax directly to PCB and connect the BNC to oscil.
~10$ DIY near probes — if the need for them arrises


Multimeter(s)
—————————

35$ UT210E — cheap “expendable” multimeter

700$ SDM3055-SC 5 1/2 — Scanner card capable multimeter bench multimeter
(Why?) this may come as surprise choice, but I find it remarkably useful to see multiple results of a voltage/current from a board I am debugging, or for longer tests. Also it is more of a future proofing for climatic chamber (in the future, and absolutely certainly DIY)

((Does anyone want to dispute this choice?))




VNA
—————————

300$ NanoVNA V2 Plus — from official seller HCXQS group
(Why?) Maybe I am too paranoid about getting those extremely cheap from aliexpress and elsewhere? Should I just go with the 100$ clone? there are sighns that they are not as precise





organization
—————————

I bought some SMD books (resistors, capacitors)
and a bunch of empty SMD books that I will make myself for common components
all of this being tracked in Excel by book, row, column numbers, and seachable by CTRL+F

[Fungus]

35$ UT210E — cheap “expendable” multimeter

Ew! Have some self respect.

Get a Brymen. At least a BM235.

[dobsonr741]

You listed consumables together with fixed assets. I’d list them separately, and consider a monthly spend on consumables, or establish supplies for a whole year. Your accountant will love it.

[armandine2]

20$ generic usb microscope from Aliexpress

you must have excellent eyesight

[janoc]

20$ generic usb microscope from Aliexpress

you must have excellent eyesight

He likely does, given that he is 24. But I agree, those $20 USB microscopes are useless crap with flimsy stands that will not allow one to do anything.

OP, as this is for actual work and not only for fun, do yourself a favor and invest in some proper equipment.

It doesn't need to be expensive. E.g. forget those "wireless" soldering irons - if you are hoping to do more soldering than attaching a wire here and there, you will regret buying these. They are utter garbage as daily drivers, poor choice of tips, poor power and expensive. All to get rid of a cable? What for? Get a cheap station that can handle genuine JBC tips and it will be the best investment you have made, e.g. one of the popular Aixun ones.

That Yihua for $200 is utter garbage too. That's a cheap $20 hair dryer attached to a very poor quality preheater. You will certainly not be able to desolder and rework any larger BGAs with that, it is simply too underpowered for it. Don't waste your money on that, rather buy a proper hot air station for not a lot more - e.g. the Quick one. If you need a preheater then various SMD rework hotplates can be bought cheaply.

Multimeter has been mentioned already.


What do you need the NanoVNA for? Are you hoping to do any RF work? If not it is a waste of money that you should invest in other tools and consumables instead.

Such as a cheap logic analyzer - debugging USB or serial buses on a scope can be done but it is a pain in the butt compared to a dedicated tool. Again, doesn't need to be expensive.

[Veteran68]

35$ UT210E — cheap “expendable” multimeter

Ew! Have some self respect.

Get a Brymen. At least a BM235.

^^^ This. You're building out a nice home lab with a really decent scope and bench DMM, but then you specify a cheap clamp meter as your handheld? I mean, sure, go ahead and get it to have for when you need a clamp meter. It's $35. But don't consider it a substitute for a good handheld DMM, and I personally consider at least one -- preferably two -- fully capable handhelds in addition to a bench DMM to be a necessary component of any lab. I probably use my BM786, BM869s, and Fluke 117 more often than any other instrument in my lab, including my two 6.5 digit bench DMMs.

I concur on the Brymen choice. I'd recommend a BM786/BM789 if you don't want the big daddy BM869s.

20$ generic usb microscope from Aliexpress

you must have excellent eyesight

LOL, also this. At least look into the ~$200 Andonstar microscopes, ideally with HDMI out so you can connect an external monitor. Otherwise I'd say a set of magnifiers would be a better investment. USB scopes tie up a PC to see anything, and the quality and flexibility just won't be there. After trying a much cheaper microscope w/ display that cost about $60, I upgraded to the Andonstar AD246S-M attached to a 24" FHD monitor and it's great for my purposes. One day I may upgrade to a proper trinocular scope, but to this point have not been able to justify it.

As to the rework station, I strongly suggest the Atten ST-862D. The Yihua rework stations are junk.

[djacobow]

I believe you can't get the scanner card later for the sdm-3055. If your want it, you need to order the unit with it. I like this dmm.

[Veteran68]

I believe you can't get the scanner card later for the sdm-3055. If your want it, you need to order the unit with it. I like this dmm.

He quoted the -SC model at $700 which includes the scanner card. The non-SC model sells for well below $700.

[nctnico]

Looks more like a hobby-grade lab that way then a professional lab. Get some A-brand gear in there like a scope from Keysight, R&S or Tektronix. The R&S RTB2004 is a good choice for doing embedded work. Same for a DMM; Keysight and Tektronix have solid performers for reasonable prices.

For soldering look at Ersa; their tips never go bad which is a huge help to get soldering done quickly. JBC and Weller have a strong following and I have used my fair share of their irons but I'm always unpleasantly surprised by how bad their tips are in comparison with Ersa. Also make sure to get some SMD hot tweezers (Ersa's chiptool for example; one of the better models I have used). Hot air is something you can go cheap on. After all, hot air is just hot air. And stay clear from cheap desoldering braid; it ends up costing more in wasted time. I don't know the MG Chemicals brand but looking at the price it is in the same ballpark of Chemtronics 'Soder-wick' which works very well.

Keep in mind that people chasing their tails and/or wasting time due to bugs / problems with equipment gets costly real quick. Both in lost hours but also delays in time-to-market. Time can not be recouped or replaced. Having good technical support for the equipment is super important. Been there, done that and learned this lesson the hard way.

It is essential to think in risk (of lost time) versus investment where it comes to tools that support the success of your business. Missing a deadline for a demo or needing more time to create a fake demo is killing for developing the actual product in time. Equipment that slows you down becomes a 'double whammy' that way.

[Martin72]

Get some A-brand gear

He did after all.

[dobsonr741]

Agreed. “A”liexpress-grade. 

[mmame]

I'm missing a - probably even pretty cheap/simple - Logic Analyzer in the list... Number of ports, sample rate and voltage range according your needs. But I'd highly recommend devices with Sigrok support.

[coromonadalix]

unlees the @OP get an analyzer with a scope   loll

well  wanting to build a shop and cutting corners at 1st,   means no seriousness

luckily  there is/was better suggestions given out there

5K$  doesn't bring you lots of good stuff  nowadays, under 10K$  you're not totally serious


Take scopes who are hackables,  same for sig gens,   buy a good iron to have peace of mind at 1st, yeah tips are @%%#

desolder gun mmm    old high power hakko series where good, for thick pcb's  and dont cut corner there, you'll get what i mean

I'm not sold to Hakko,  but got good tools from them,  solder iron, smd desolder tweezers,  hot air tools

Siglent scope hacked, hf Siglent gen hacked, uhf Siglent sig gen hacked  ........ saved over 2.5k in unlocking options, Siglent Spectrum analyzer ...  and that's well over 10K just for Siglent stuff

A picoscope for a mere 3400$  usd   outch  this one hurt

i do have an Rigol scope hac... and Rigol Spectrum too,  but they became second pretty fast, because of lack of simple to use software(s) for pc interfaces, and a big lack of labview mathlab  etc ...

i'm already over 25K $$ 

aaaaaaaaaand the job bought an used 7254b  for the UHF experiments, but already at limits, we bought this ##%#% of 7404  to push the envelope further away  loll  to be controlled by pc

Even some china  gimmick  like 3x Juntek jds-2900 40Mhz sig gens, simple and efficient, pc controllable

Chat GPT was kind enough to slightly rework the code for a self developped software piece, not impacting Excel (keyboard macros etc ..) when these 2 runs at the same time for the juntek  sig gens  lol

EDIT

I do use some usb microscopes with an integrated 7" lcd on a tripod,   really good definition and light  for 70$ usd  loll
even one usb at 200x i run every where on a pcb for 30$ enough precision at 640x480 loll

[tautech]

1200$ SDS2104X (with maybe slight fw modifications) — All the features I need for 95% of my measurements, good sample rate, good BW, signal generator

Do consider getting the current promo bundle with it:
https://int.siglent.com/info/detail-74.html

[DaneLaw]

35$ UT210E — cheap “expendable” multimeter

Ew! Have some self respect.

Get a Brymen. At least a BM235.

It's a low-current AC+DC clamp meter, I doubt the Brymen BM235 will do jack in that regard..the UT210E is still a very good choice if you looking for a low-cost 1mA res DC clamp meter.
UT210E is quite decent in the low DC amp, not least with the price in mind to around 30 to 35 bucks delivered. - HKJs review https://lygte-info.dk/review/DMMUNI-TUT210E%20UK.html

As a general multimeter.. a clamp meter is far from ideal, but since OP highlights a low-current AC+DC clamp meter, I would guess the specific low-current DC clamp ability is important to OP
- as otherwise, you would not pursue a clamp meter that lacks fundamental features (freq/hz etc) as your DMM.

Dave looked into stocking up in low-cost Brymen AC&DC clamp meters' as Brymen could offer them at a very attractive price, but I don't think he followed through?
- as viewers posted' it would be a hard sell versus these cheaper Chinese versions with overall better low-amp DC tolerances/performance - to a fraction of the cost
https://tinyurl.com/yckasaw8 - also took a one-to-one look https://tinyurl.com/mr3naaf3

[mmame]

I don't get it: why should spending below 10k USD be "not serious"? As we don't know what he's going to build, we can't really tell him how much hwe "shall" spend.

I was in a similar situation around 12 years ago: a SW company but we needed some "simple" hardware for a SW project (low voltage and current, low component count, no RF stuff...). Bought a Rigol Oscilloscope for less than 1200, lab power supply was an ancient selfmade one from home, prototype PCB's were soldered in an old frying pan...

Of course, i.e. for EMI pre-tests, we had to go to a lab but finally everything was up and running - and certified ;-)

As said: it depends on the specific requirements.

[Martin72]

What I imagine a "laboratory" to be:

- Oscilloscope
- Spectrum analyzer
- Signal generator
- Multimeter, several
- Supplies, several
- Loads, multiple
- Soldering station
- Rework station
- Reflow oven
- Microscope
- Thermal camera
- Current clamp, one stand-alone, one for the oscilloscope
- Tools in general, tweezers, pliers, circuit board holder, etc
- Test leads
- Work lights

If you can manage all that for 5000, respect.

[nctnico]

Agreed. When I went from a job to being self deployed and needed my equipment to make money for me full-time, I quickly found out that my lab was not up to the job. I needed to make some serious investments to get projects out of the door in a quick & efficient way. For essential tools like oscilloscope, PSU, soldering iron, hot air and DMM make sure to have spare/extra units so you are not in a jam if an essential tools fails.

Although I would scratch the reflow oven and rework station from the list. Doing in-house PCB assembly is not a good idea from an investment (equipment + time) and quality perspective.

[Martin72]

We have something like this in order to be able to quickly implement any changes to externally assembled circuit boards, of course you have to calculate what is worthwhile and when.

[PlainName]

I have an unique opportunity to get “tabula rasa” lab. To build it from scratch. I will list what I think of each instrument and why I decided to go with it.

I would suggest that you get the specific stuff you know you want/need and that will work for you, and the other... well, cheap means you buy twice, but if you buy the wrong expensive tool you're stuck with something you don't like and is hard to replace. So I would buy something that seems reasonably capable but is cheap enough to lose, and from that find out what you really want.

For instance, if you were buying a thermal image (which, BTW, is good for bringing up prototypes, not just repairs) one of the features that isn't obvious to a new user is the ability to manually set the temperature range. Most cheap and mid-range TICs don't have that but get recommended because of.. well, there must be something that appeals to the opinion provider. Get a cheap one and find out these things like this, and when you get the expensive replacement you know exactly the feature set you want.

Similarly, I've bought recommended expensive kit and regretted it because something is the way someone else likes stuff, not me, and it sits unused.

Your microscope may be a good example. I know some people manage with them (and, after all, Alex at Northridge uses just a pimped one), but the only way to find out how bad it is for you is to try it. It's cheap enough to do that and I bet you'll pretty soon go for an optical one, and a Barlow lens would be a feature you need.

The downside of going this route is that it takes a bit of time to figure out what's good and what isn't, so rather than diving in and hitting it running you'll be slow-starting every new thing you're doing. But you'd need to learn to use the tools (and which tools) anyway, so perhaps there's not that much difference.

[nctnico]

Similarly, I've bought recommended expensive kit and regretted it because something is the way someone else likes stuff, not me, and it sits unused.

Top tip:
When buying something expensive: 1) do research to understand pros/cons 2) create a test plan 3) get the device on loan for a thourough evaluation (or make sure you can return it).

Over the years I have send quite a few pieces of equipment back because it didn't meet expectations. There is nothing wrong with that.

But I agree with your bottom line though; in the end it is only yourself who can decide which piece of equipment fits best.

[ebastler]

What I imagine a "laboratory" to be:

- Oscilloscope
- Spectrum analyzer
- Signal generator
- Multimeter, several
- Supplies, several
- Loads, multiple
- Soldering station
- Rework station
- Reflow oven
- Microscope
- Thermal camera
- Current clamp, one stand-alone, one for the oscilloscope
- Tools in general, tweezers, pliers, circuit board holder, etc
- Test leads
- Work lights

If you can manage all that for 5000, respect.

For the type of projects outlined in the original post, I think the spectrum analyser and electronic loads are not required, and probably the current clamps as well. (Get one of those USB power meters instead, FNIRSI FNB58 or similar.)

I agree with nctnico that a reflow oven is also not a necessity. But a rework station (at least a hot air soldering iron) certainly is, and a minimal version was included in the OP's shopping list.

For a microscope, personally I prefer an optical stereo microscope by a large margin. If you don't expect to sit in front of it all day but only for occasional checks and repairs, the slightly uncomfortable sitting position when looking through the eyepieces is not a concern; and the stereoscopic view helps a lot! Get one with a large free working distance, to be able to solder and manipulate components comfortably.

As others have said, get a wired soldering iron for the bench. If you really find yourself doing "mobile soldering" a lot, you can add a battery-powered iron later.

I don't concur with the "get an A brand, anything else is unprofessional" attitude. But I would strongly suggest to get properly safety-certified equipment in this context, to meet workplace health & safety and liability requirements. The YIHUA rework station might not make the cut here.

Overall, I don't see anything fundamentally wrong with the $5000 target. Electronics work is a sideline of the planned business; and if you want to start growing the business incrementally, rather than selling your soul to an investor or a bank right away, you need to be thrifty.

[nctnico]

For a microscope, personally I prefer an optical stereo microscope by a large margin. If you don't expect to sit in front of it all day but only for occasional checks and repairs, the slightly uncomfortable sitting position when looking through the eyepieces is not a concern; and the stereoscopic view helps a lot!

A good quality magnifying lamp can also do wonders. These take a minimal amount of desk space, can be looked through with both eyes so you have depth perception and serve as a desk lamp as well. I have one from Daylight with a 5 diopter lense and this is good for soldering 0402. For closer inspection I have a seperate magnifier. The advantage is that this allows to look at a board at many angles to check solder joints. A joint may look good from the top, but looking at it sideways may reveal things are not OK. Even looking under BGAs to get a glance of how well these are soldered is possible.

In the past I have used a Mantis as well and I liked it (with 4x or 6x lenses) but I don't have the space for it. Using a stereo microscope doesn't appeal to me though. And microscopes typically have too much magnification for soldering work so you'd need to get extra lenses that reduce magnification and increase the distance between lense and board.

[dobsonr741]

If I would co-founded a startup and I’m the chief hardware EE, I would not buy all the gear I can think of blindly without the actual need these day. Everything is just a click away, I would only by a DMM and a scope (Rigol, Siglent or a Tek), a reputable and reliable soldering iron and nothing else. If anything extra is needed you can always have it within days, meeting or exceeding what you need. And not from Ali, as this is an investment, assuming value kept and returned when used. With Ali you have no warranty, no specification to count on and can fail at any time. All that nicely devalues your investment. Still can be appropriate, but do not count on them, it’s a one time disposable asset. Also priced by Ali/CN manufacturers that way.

[Construct]

If I would co-founded a startup and I’m the chief hardware EE, I would not buy all the gear I can think of blindly without the actual need these day. Everything is just a click away, I would only by a DMM and a scope (Rigol, Siglent or a Tek), a reputable and reliable soldering iron and nothing else. If anything extra is needed you can always have it within days, meeting or exceeding what you need. And not from Ali, as this is an investment, assuming value kept and returned when used. With Ali you have no warranty, no specification to count on and can fail at any time. All that nicely devalues your investment. Still can be appropriate, but do not count on them, it’s a one time disposable asset. Also priced by Ali/CN manufacturers that way.

Having done the shoestring budget startup thing before, I fully agree about buying things as needed.

Budgeting is good to calibrate your expenditures now, but it's generally a mistake to rush out and buy a lot of gear you think you might need in the future.

AliExpress can actually be fine for certain pieces of gear if (and only if) others have already vetted the supplier and product. Don't go the AliExpress route for mission critical gear, though.



Keep your priorities in mind. A 5.5 digit benchtop multimeter is very convenient, but you can't justify it as one of your first expenses because you might use it with a future climate chamber. If you need it for the future climate chamber, buy it with the future climate chamber.

For the rest of the gear, it's really hard to say without knowing what you're building. Is the NanoVNA something that you think might come in handy in the future, or does your work require a lot of RF measurements? If you're doing RF work, I'd start cutting budget on other things (5.5 digit DMM, Sensepeek setup) to prioritize everything you're going to need for that. Cables, adapters, and other gear add up quickly. If you're not doing RF work, I'd skip the NanoVNA entirely and wait until you have a specific need, then buy the specific tool that will solve it.

Don't spend all of your budget up front. You will encounter unexpected tool needs as you go along.