EEVblog 1415 - Reverse Engineering the DP10007 Differential Probe
Vložit
- čas přidán 4. 07. 2024
- Reverse engineering the Micsig DP10007 high voltage differential probe.
Turning the PCB into a schematic.
Reverse enginering how-to for the Rigol oscilloscope: • EEVblog #675 - How To ...
DIY light box for PCB photography: • DIY Product/PCB Photog...
Forum: www.eevblog.com/forum/blog/ee...
Subscribe on Odysee: odysee.com/@eevblog:7
EEVblog Web Site: www.eevblog.com
The 2nd EEVblog Channel: / eevblog2
EEVdiscover: / eevdiscover
Support the EEVblog through Patreon! / eevblog
AliExpress Affiliate: s.click.aliexpress.com/e/c2LRpe8g
Buy anything through that link and Dave gets a commission at no cost to you.
Donate With Bitcoin & Other Crypto Currencies!
www.eevblog.com/crypto-currency/
T-Shirts: teespring.com/stores/eevblog
#ElectronicsCreators #HowTo #ReverseEngineering - Věda a technologie
I think the dual n/p channel transistor arrangement is actually an h-bridge for switching the _latching_ relay. Cool videos! Love that stuff!
What's the point? A single transistor does exactly the same job. This is not a mechnical latching relay.
@@EEVblog AGQ210 is 1 coil latching relay according to the data sheet.
Haven't read the datasheet, but it is listed as latching relay on mouser and having worked with bistable latching valves in the past, I assume that you have to change the direction of the current flow inside the coil in order to switch it.
@@davidmenasce6614 Ah, so it is, I missed that on the datasheet. The AGQ200 is the single side stable one, 210 is latching. I don't see any reason to use a latching relay here, it's not like power consumption is an issue.
@@EEVblog Me neither. Maybe a relic from a battery operated version or they had plenty of these relays in stock?!
The relay is AGQ210A03 which is 1 coil latching, so relay keeps position without power. Therefore you need bipolar driver to reverse voltage for switching.
Yes, I missed that on the datasheet. The AGQ200 is the single side stable one, 210 is latching. I don't see any reason to use a latching relay here, it's not like power consumption is an issue.
@@EEVblog Maybe they decided not load the switching converter with the 100mW of the relay all the time, either heat up or noise in its output, or that output voltage goes "too low" with the relay load.
The PCB capacitor is for matching the positive and negative input(impedance). If the capacity matches you don't use them. The input with the lower capacity will get additional PCB capacitor for best match. High voltage capacitor have higher tolerance, so you might need this matching mechanism.
I just use GIMP. Add layers (one reversed, of course), set the alpha channels, then toggle the visibility of the top layer as needed (eyeball icon on the layers toolbox). You can even set a shortcut key to toggle the visibility of the *current* layer. Not perfect, but better than printing stuff out...
I trace things out with GIMP. Either photo or scanner works (if the board components aren't tall it'll be in focus). I draw lines on a layer all white. Any lines that may intersect, I put on a new layer. Then I can flood fill paths color coded [red = 5v, yellow =12v, etc], and merge them as i go. Doesn't take "that" long depending on board complexity.
I use this same technique, although with Photoshop. It works amazing when trying to trace through a circuit. I usually do it on my surface pro, which had a stylus. I can then add layers to draw lines on the traces to show the exact path of a signal or power supply trace.
Someone reverse engineered a Kindle that way, there's even the GIMP document available to download somewhere.
In any photoshop-like tool, the magic wand can be very useful for following traces. I adjust the levels of my board so that there is good contrast between the traces and the bare PCB. Then to select a trace, I just click it with the magic wand, and I get the whole trace. Then I tint the selected area with colour to both mark it as “done” and to indicate function (i.e. power, logic, analog, etc.)
If you have a wacom tablet, it's even easier to do the reverse engineering in GIMP since you can actually draw notes and diagrams.
Just a comment on getting good images for reverse engineering. About 15 years ago I was doing a first article inspection for a power supply I had designed after I received it from the factory. I figured out that instead of sitting in the lab at a microscope to inspect solder joints and looking for flipped/rotated components, I could just scan both sides of the board on our copier (at high resolution), and then I could inspect the board at my desk by zooming in and viewing the scanned image. A big advantage of "taking a picture" with a flatbed scanner is that every part of the board is seen as if you are looking at it from directly above (which of course is how the image is captured by the scanner). Scanning an assembly would give you the same advantages when trying to reverse engineer a design, and it would also enable the images of both sides to be at the exact same scale.
How about starting a series about Reverse Engineering. ?
I would like to see videos like this :D
Not a bad idea, at least one video could be filled as there are different ways to reverse engineer a Pcb depending on how complex it is.
@@IanScottJohnston ya true
Hell yeah.
I've done several other videos on this already.
@@EEVblog - great. Now do more! :-)
"TI 98 OPMI" is an OPA2171 dual CMOS opamp.
There is an auto-zero function. If you press and hold both x10 and x100 buttons at the same time, it zeroes the offset. So the microcontroller is doing a bit more than overvoltage detection.
hi the reason of the 2 6604 fets is that the AGQ210 is a latching relay single coil so I guess they're here to invert the voltage across the coli to make if change state
FETS configured as H-bridge.
Dave, AGQ210A03 is a 3V single coil latching relay (relay is set and reset with different direction coil current). To drive a latching relay two complementary mosfets are required configured as a H-bridge, so the 6604 is a N- and P-mosfet pair. Latching relays are often used in battery powered applications .
This brought back a lot of memories. Back in the early 80's I did all of my own board layouts (for logic card test equipment) using the forward version of this - i.e. acetate projector copies of top, bottom and 2 internal power layers in different colours sheets then using foil pens to add the connections - great fun. Passed the roughs to the CAD team who would tidy them up and send me back proofs - back to the photocopier with colour acetates and check everything out before having them sent out for etching.
Another trick for reverse engineering is to put the board on a light panel to shine light through it as it makes it a lot easier to see the tracks, I use one of my flat studio lights, on its back and just lay the board on top.
Great idea! Thank you. Cheers
Mark
Yeah, mentioned in the video with a text overlay as well.
@@EEVblog So you did, right after I commented... couldn't be bothered editing it :-)
@@TheDefpom 'Couldn't be bothered' seems to be one of the recurring themes of this video :-D
@@pnjunction5689 Hahahah !
We have several (15 maybe ?) of these probes. We do have a few that developed a DC offset that can not be calibrated out by pressing and holding the 50X and 500X buttons. This video might help us to fix those ! Thank you Dave ! 😁😀
Yes, I'm you're uncle ! (bob)
I bet that the extra circuitry has to do with the microcontroller's role in adjusting the offset when you short the input probe and hold the two buttons down for a few seconds. A/D input, being slow-ish, would also make sense for that role. Read the offset and inject another offset to counter it.
That is what I would probably do anyway
I opened a question on the forums just a couple weeks ago about alignment of PCB images. Usually in reverse engineering, you have the boards on hand to take pictures of. But not always.
I still have a flatbed scanner and use it to scan PCBs for reverse engineering/repair etc. All you need to do is rotate and align them, no lens distortion. Components shouldn't be too high or you do get some distortion, so mostly good with smd boards or bottom side. Scanned images also have a set DPI so you can do 1:1 printouts if that's needed and you can measure distance in software without having to convert number of pixels to measuring units. There's also panorama/stitching software like Hugin (open source/free) that can align images automatically based on a few points you give it. Or deep stacker software that can do the same if you don't want to do it manually. Hugin can also remove perspective and lens distortions.
I use Photoshop for alignment and Illustrator for annotations because it's vector based and is easy to scale/edit/change. But Gimp and Inkscape can do the same job, just a matter of preference and availability. I usually also change the colours between top and bottom layer for easier tracing out of vias etc. using the transparency slider to switch back and forth or using the layers panel eye on/off button.
❌Chapters Added
The wisdom of videolytics cannot be matched
Nice job!
About these dual transistors - I didn't check that relay, but one option for this two dual transistor schematic design may be, if that relay is with latching option and MCU changes polarity for remote..
@15:15 I'm not so sure that's pin 18 Dave ;)
IC TI OPMI with VSSOP package is OPA2171AIDGK 36-V, Single-Supply, SOT-553, General-Purpose Operational Amplifiers
@3:55 I use an ancient Paintshop Pro to do just that for reverse engineering PCBs. I start layers for the front and back of the PCB, then I add layers for front and back traces, which I trace manually, often in different colours depending on the function, for instance data lines, power lines, control lines etc and another layer for labels, mostly the componenet numbers.
I find it very useful but I have always wanted to make something more dedicated and have worked on and off on a custom application to do it in a more streamlined way though haven;t progressed much. In the meantime Paintshop Pro is proving quite helpful. I guess Photoshop could also be used but I found it very cumbersome for this sort of thing.
Resistors in series also add terminal voltage.
Helpful methodology, thank you! In terms of doing this manually, it seems like an image editor capable of layers (like Photoshop) could be hugely helpful. Just use the contrast or transparency controls on the layers, and mark up layers above. I'd be curious to see how the gentleman you pinged on Twitter automates this process.
I think the feedback from output is for automatic zero adjust voltage and over-voltage detection.
My instant thoughts exactly. I wouldn't be surprised if the MCU has a tap somewhere at the input of the cable driver to null the offset at idle.
How can it know when its idle? Sometimes the equipament is turned on when there is voltage being applied to the input already.
@@andrelange9877 Mayve when the signal is below a certain level for a certain amount of time it assumes that it is DC offset. Probably when one want to actually measure a constant DC voltage it won't choose that low measuremrnt range that the signal is lost in the noise.
Or maybe it expects a shorted input at power-up or constant hold of a button or something else to calibrate itself, I haven't read the users manual of this exact device.
Applying test signal to an unpowered instrument is a bad practice already, so when it just happebs that it skips calibrating itself in that case is just the best outcome. It can even use the previously saved offset if there is an EEPROM.
@@gabiold The first paragraph would present a very annoying and possibly dangerous strategy. Sometimes one needs to measure small voltages like 5V or less and a differential probe is used in one channel to prevent a ground loop with the other scope channel that is connected elsewhere in the same circuit. Sometimes the differential voltage is small but it sits on top of a high potential line, so the differential probe is also needed to protect the scope input. A small voltage reading is a small voltage reading. It cannot be auto zeroed out just because some clever logic thinks that it *may be* uninportant information and *likely* an internal offset. Not so likely.
@@andrelange9877 Measuring 5V with a 200X range HV probe is not very useful. And if the software knows that the probe could have uV or mV offset at max, it wont compensate for 5V.
And we just guessing here without even knowing what the MCU measures. Quite possible that it sees the output of both the input buffers which can be useful to determine where the DC offset is in the system and to see that both probes hang on an 1000V common-mode signal.
There is at least a dozen ways how it can be done properly...
2:26 That's handy. I have been using transapency tool "peek through" by Luke Payne with two Irfanview images one flipped underneath for that sort of thing.
Old fashioned 'Paint Shop Pro' allowed adding multiple graphics layers and geometry layers for text and arrows which you can adjust the transparency between them and their layer order and layer-layer alignment on the fly. In fact, I would use and still use today my old Paint Shot Pro 6 from the year 2000 to do exactly what you want. It also has the edge outline filters if you want such image processing. Also, if you have equal spacers at the 4 corners of the pcb, some flatbed scanners will generate a perfectly square edge-edge accurate image even though the spacers lift the PCB off of the glass so that the components don't sit in a bumpy random height from the flatbed scanner.
Hehe first ting i thought, maybe Paul Daniels can make a program, and tada, he is on the job!
18:58 It's a latching relay so you need to be able to reverse the polarity on the coil.
Edit : wrote this while watching and saw that it was already mentioned ;)
5:17 The proposed tool could easily do that too, i.e. align the two board faces, magnify, de-magnify, stretch, skew and remove lens distortion and shading as necessary all in one go.
I usually take off all components after taking multiple hi res pix of the fully assembled PCB... and then put the PCB without the components in a flat bed scanner and scan both sides. Simple !
I wouldn't call taking off all the parts from a PCB "simple"! But yes, if you were super serious about fully reverse engineering and producing a accurate clone or whatever, this is what you would do.
@@EEVblog Yup... Done multi-layer boards like that... sanded the top and bottom solder mask after scanning and then etched away the top and bottom layers.
Went the local hospital for X-RAY. Scanned the X-RAY in the flat bed scanner and .... as you say... "bob's your uncle" :)
Its a latching relay which needs to be reversed to unlatch ;)
I think the uC analog inputs are setup as comparators, not ADC. That micro will not the bandwidth to sample the signal to catch the maximum peak, but the comparator will; otherwise you would need a peak detector circuit in there.
Upload the two PCB images to github as two versions of the same file and you get an image difference view with a bar you can drag to swipe between them or fade between them.
Q6+Q7 could be an H-bridge configuration for powering a bistable relay.
Stuff around the microcontroller looks like it samples the output voltage, and then it is possible to communicate via the USB interface, to software that can use the data, things like range, clipping status and likely also drive the output to a defined voltage, reading it back via the sense after the resistor, to accurately see if termination is on or off. Otherwise just having the inverting stage you make clipping on the negative half cycle detectable, as the internal ADC of the chip is probably unipolar, so needs the inversion to detect negative clipping.
Also the ability to read output allows sensing offset voltage and nulling it, either during manufacture, so the micro can put a defined small voltage as bias to the output, or during use if a null is desired.
There is no USB data isolator, and this is not a USB enabled micro. The two lines from the side USB connector are only for basic port charge setting resistors.
Great video Dave. I wish you would do a video explaining how Spark Plug probes work. Once I tried to make myself a very high resistance attenuator to read a spark plug ignition signal. I almost blew my digital scope input stage. The design was very straight forward. Some very small caps and high value resistors. For some reason my digital scope would go haywire. Never could get a decent signal. My guess is such high voltage signals would need more than just a couple of pf caps. Maybe an idea for a next video. Thanks
Perhaps there's too much noise in the hv pulse?
A totem pole switch on both sides of the relay? I converted some standard relays once for use in a security system with a security DVR. A custom box to switch cameras in and out. A custom 12V UPS. It contain more than 16 power hungry relays. However, I opened all the relays and placed hand matched per
The spare opamp and parts are probably a peak detector, I doubt the micro adc is fast enough by itself
Maybe it is a latching relay, then you'd have to build a h bridge or something to switch between states
Yay!
How to reverse engineer microcontroller/power supply section:
1: Type in part number of chip in google
2: Hope reference schematic is available somewhere
3: Usually its an exact copy because engineers prefer not to waste time
It is a coil lock relay! for low energy consumption
I just purchased this unit. Like it so far. The only negative for me is the use of male banana jack connectors on the probe (for high voltage safety I guess). Which makes it practically impossible to find other test lead accessories to plug into it. Every single test lead I find has the male jack on the leads, so none plug into the probe. Has anyone found any needle point test leads (like on multi-meter) that plug into this unit in place of the honky ass test leads they provide?
The worse thing I've tried to reverse-engineer so far is a solar charge controller. Had to remove some current-sense resistors to make sure I was putting components on the correct side of the sense. Got enough of the schematic together to try to find out why it wasn't working, but then it suddenly WAS working again?!?! Of course now its water protection coating has a bunch of holes in it due to the continuity probing.
Biggest problem with auto-reverse-engineering though is multi-layer boards. Even single layer can be a pain due to traces running under components, but when there are internal layers you pretty much need an ultra-high-resolution CT scan to locate and reconstruct the trace paths.
Dave can you please do a video on MOSFETs and Spirito effect?
What happens with PCB compensation when ambient humidity changes?
The second output feedback signal through the op amp - we don’t know what C53 is, nor the second half of the circuit, but I’m guessing this is a filtered version (maybe rectified if Q5 is involved) for whatever reason. Could this be used by the auto-zero mode? Also since this 2nd feedback comes from the output side of the series source resistor, the software could perhaps determine whether a 50-ohm or a high-impedance load was connected, but to what end?
Should a coax cable be attached to the board in that way? Will that 1cm of unshielded coax not cause an impedance mis-match? Maybe a problem at higher frequencies? Just half a thought anyway..
You can load in images into KiCAD's schematic editor and draw traces on top of it. I have done it before for a couple of boards but it is a bit clunky. I think it would not take too much effort to vastly improve. What would be a HUGE help is a feature is a toggle to lock the net. I want to be able to freely grad components and wires around with out them automatically connecting if I happen to touch a wire with a pin, or two wires overlap each other. It would also be great to be able to freely stretch and scale components to match their physical size on the image. You would be able to snap them back to their original size once the net is locked.
1) Prepare a high contract image of the front and back and reduce the resolution and bit depth. You need a small file size since KiCAD starts to have a problem otherwise. Less than 1MB is good.
2) Put a mark in the center of the image and put a cross hair on the schematic so you can easily swap front and back and maintain alignment.
3) Put the images in first since KiCAD draws objects on the screen in the order they appear in the file. (You can fix this if you open the .sch file in a text editor and copy/paste the image block).
4) Turn off orthogonal wire mode and just follow the traces around, add component symbols as necessary.
5) Move the image off and start dragging components to better places, shorten and simplify the wires.
Mr. Woo is coo
Also u can use Photoshop software by opening the two sides of the pcb and then using opacity feature to fade one of those layers, i think 🤔 it's a 👍 idea 💡 ok bro, anyway thanks for every video you post keep up bro , from north Africa 🌍
Target Layout Software has this "Reverse Engineering Thing" built in ... They even offer a free Version !
A schematic for PWM constant current source do you have?
How come ADC samples e.g. sine wave right from output of transmission line driver opamp without voltage shift to positive side?
With relatively small number of parts, why isn’t this differential stuff incorporated into all scopes? 🧐 It would be nice to have this capability.
Better question is why these are so f'kin expensive? Practically nothing that special, not even any RF magic apart from the extra ground layer.
4:21 those programs you say exist . An augmented reality based software that marks traces and components . The name of the software is inspectar
If the tool is any good I woud buy it today.
Were you in Redmond Washington about 3 weeks ago?
I don't think I could see where and how you determined the front-end op-amp power supply and single ended bias offset ?
Trying to fix the offset issue with one of these Micsig differential probes. The most annoying thing about these probes for us (we have many of these probes !) is that a lot of them develop a voltage offset that is not able to be calibrated out by pressing the two 50X 500X buttons at the same time which is normally the way to do that.... I think that the micro may actually put out an analog voltage that offsets the circuit ?
Guessing on that.
where do i get image effects tools in video?
I'd be very interested, how the Input Network is EXACTLY looking, because here some values are still missing (no front on you here, i'm very pleased with your work).
Because i made a Diffprobe with the THS4631 and a THS3091 as output buffer.
I had a very interesting effect during the set up
I can not set the input network (which is just 1Meg//1p5F 110kR//[variable]pF in my case because i do not need a high common mode voltage), to a straight behaviour.
There is a point, at which the step response is still too high after 100ns, but underhangs its final value after 1us.
It seems, that the input capacity in the THS4631 is not straight. It is more something like 4pF at first, and then, behind 1 or 10 kOhms or s.th. like that, additional 6pF
(which resulted after a lot of LTSpice simulation and reverse engineering work)
For all spinoff measurements, i shorted the negative side absolutely to ground with 0R, so i started with just the positive side.
Here, the input networks are also not straight. They often use resistors in a row to their capacitors. And for me it looks like they knew, what i am looking for and solved it this way .....
Greetings from Germany,
stonix015
Oi, Mangel, chuck me another shrimp on the barbie 👍
I don't think I've ever heard "Couldn't be bothered" said more. Then sometimes going back and bothering anyway...
Gimp does a great jump of reversing lens effects and perspective. We did that with photos in combination with CT scanning the multilayer boards.
Getting the board flat for the CT was hard as you can end up with 2 layers on the same scan plane. We didn't have the 3D S/W to try and fix that. Each PCB creates a massive database of 'voxels' - volumetric pixels.
Interesting as always Dave :-), nice video BTW - when can we order the Ucurrent Gold version again? - this nice piece of equipment is always listed as "Sold Out"
Best Regards /A
Do you know how to fix a Sunbeam Coffee machine? they have some 10 year old ones on Ebay and I think your viewers would find that interesting.
hey someone in china is working on your suggestion...???? thanks great video...:)
Interesting to see how this HV probe trying to trim out CMRR imbalance in the divider string. Typical the divider frequeny response is by Rdc, Cmid and Rhf matching. They are using 2 C-trim on each side to capture also part of the Rhf mismatch.Because a Rhf-trim have too much L parasitics. A dirty trick is to avoid 8 parallel Rdc||Cmid which result in 8 mismatched time constants. They take the risk then of high voltage instability of the Cmid caps or need voltage resistance for each which increase parasitic L.
Btw all manufactures are obsessed by the measureable CMRR and forget about source impedance mismatch. In principle for 60dB rejection you have to specify the source impedance mismatch and if you know about this you will be surprised that a 2pF vs. 20pF load will be more important than 60dB. Isolationprobes where a third shield connection bypass the transient CMD current are sometimes the only solution to measure drivers in HV PWM bridges.
A tip: why not reverse eng and add the A, B, A-B function and kick out the relay.
Pain in a$$ is an understatement sometimes
This probe has 100MHz bandwidth in both 10X and 100X configurations. How is this possible, since BW is degraded for higher closed-loop gains? Or is this specified for the worst case?
Please Reverse engineering cheap 200mhz ocilascope .❤❤❤❤❤❤
Dave why didnt you mention the guy on twitter who said he made the program already and gave a github link? Doesn't it work well?
Those are not 3 diodes. The bottom one is U8.
Target3001 can do the reverse engineering you are talking about.
Hm jeah, why could it be advantageous to disconnect both sides of the coil in the relay .. I feel its in the tip of my tongue
U8 below the 2 diodes looks like a 30xx voltage reference ... 0.2% boy, seems unnecessary
Hey Dave, that software you describe sounds fairly basic. I'm considering starting it as an open-source project, since everyone has their own ideas about what features would be useful.
Tip: put one layer's edge outline in blue and the other layer's in red. Makes jumping between layers by eye much easier. Either with printed transparencies, or with layers in GIMP. You can have both visible at the same time then and still follow traces across layers.
Also, Irfanview rocks :)
Don't you call David basic. I don't believe your smarter...
@@a51mj12 I didn't call him basic, are you even literate?
As for smarter, I'm sure you're at the top of the bell-curve...
be brave dave yure in safety we arnt
Paul Daniels, are you listening? You should make this software
Without the schematics part, this is more or less exactly how I look at PCBs using photoshop...
I use photoshop for reverse engineering, I can transform pictures and align vias and pads!
This reverse engineering sounds harder than actually designing the device itself…
Thankyou Twitter because CZcams continually fails to send notifications.
fold it down completly
just use the opacity slider in photoshop or gimp whatever
The Dave character is nice but it should be more aussie and more bloke: a crocodile dundee's hat, a negative feedback shirt, a bttf walkie talkie and a sweet cheap ass DMM :o)
I know quite OK software for reverse engineering, but I can't put a link in comment. It is automatically deleted.
I am pretty sure you can do all of that in Gimp, or photoshop if you have it. Gimp has a bit of a clunky UI, but I think you could set up a reasonable workflow if you try.
Video is great, but in all honesty, that "consume style" picture of you is "a bit" cringe.
Achievement Unlocked
@@EEVblog hahaha! Happy to serve sir!
I hope you don't mind me following you love your photographs very beautiful.
Like the cartoon avatar character, you need it as your logo on CZcams
Oh Dave, please don't put cartoons of yourself over thumbnails.
I like you because you're NOT like everyone else
I did a poll, and less than 10% of viewers didn't like it. That's as good as it gets in the world of community consensus.
I Dav. , I just give a thump up but I realized it, is in your screen and not my screen! "I click it and nothing happen". so we need a tech. which can handle that, by communicate my thump up to that specific device and click the like button.
what high voltage reverse your pigs
it all over the net but not here
So how to avoid copycats? What to do when you spent hours and hours and months to design something just to be copied?
In your local country? Patent the design. Chinese copycats? You accept that they will always be able to make it cheaper and faster lol
Make copying the hardware schematic useless without also copying the firmware. Read-protect the firmware. Develop software control algorithms that relies on custom parameters that need proprietary complex offline worksheet tools to be re-calculated/re-calibrated if possible for each individual unit.
PS: I didnt mean messing with encryption, soft/hard keys and stuff. I mean if your design has some filter or logic for instance that has 10 parameters and you could either do it with 10 resistors OR with 10 software routines, do it in the software rotines. Because everyone can read resistor values and buy the same resistors but few can unprotect and read (disassembly) MCU/DSP code and understand why your program works and theirs doesn't.
Pot the whole thing 🤣
This is a text book design anyway, to simple to worry about but I guess it's very annoying.
If the curcuit is that simple to design and build then why it took so long for these probes to appear on the market with Chinese brand and affordable price?
Micsig brought some hope for me. I'm just starting to develop power conversion products in my own business and would need to rent a space in an university lab for a limited time only to be able to validate prototypes using high voltage probes and high bandwidth current probes. Now I can have all that in my lab.
Background: In this country (Brazil), TekTronix, Lecroy, Keysight etc are prohibitively expensive for a small business or individual use, mainly due to currency conversion and taxes.
how fucking much money do you need
Photoshop
oh my, TALKS sooooooooo Much says so little plz no.
You must be new here, welcome!