undefined | Better HN

0 pointsmiahi9y ago0 comments

A one-channel scope is useful, but it still feels as cutting your left arm when you are used to (and need) more than one channel. And these days most of the things are digital, so that means SPI/I2C/whatever with more than one line, and even then, A-B is quite important. I guess you can't combine two of them with a hardware trigger in/out so you can synchronize them and simulate a two-channel.

As it is, it's more like a toy scope, fortunately pretty close to impulse buy prices as these things go, so I guess it will be alright. I would buy one but it's iOS only, so I'm stuck with my Rigol :).

0 comments

blackguardx9y ago

You are very perceptive. We had a prototype of two scopes synced with a cable we called "nunchuck mode," but we shelved it because we weren't sure how useful that would be.

I think the main selling point is the portability and remote measurement aspects. It fits in your jeans so you can literally have a scope with you at all times. You can also capture waveforms from over 100 ft away.

I was worried about the one channel limitations when we first decided to go with this form factor. After thinking about my own scope usage, though, I realized that I use only one channel on my four channel scope 90% of the time. More channels is obviously better, but not always necessary in every instance.

Thanks for the feedback.

ChuckMcM9y ago

First, I think you've made a really creative project. You did a good job of laying out its features and what you saw as the market need. The price is certainly inline with a number of other scopes in its range.

EE Times did an excellent survey article on USB oscilloscopes the under $200 ones are here: http://www.eetimes.com/document.asp?doc_id=1326699&page_numb...

There are two serious challenges with "remote display" oscilloscopes, and that is connectivity to, and compatability with, the display. Over the the years I have owned a couple but the challenge is always that when you buy into a remote display'd test tool you bet that both the manufacturer and the environment will continue to exist. For me I've got a nice 2 channel 100Mhz USB scope that has a "display and control" application that runs on Windows 98. It kind of worked on WinXP, and barely worked at all on Win7 and works not at ll on Win10. It also uses USB 2.1 which, from a connection standpoint has been exceptionally long lived. The previous "universal" connection was parallel ports and those went away and a lot of gear became useless.

The choice of IOS is ok for now, but Apple will rev IOS in an incompatible way and if you're gone by then the scope is dead meat. Compare that to my first oscilloscope that was a used Tektronix 465b that was built in 1980, recalibrated in 1990, traded to a friend in 2000 and is still running today and doing the job. Test equipment lives a long time because the job is the job, it doesn't change, and if the tool is self contained it will never not be able to do the job until the parts it uses are no longer made.

So as a tool buyer (and I'm an outlier, having bought the 465b, a Rigol 1152D, a Tek 2216, and then a Tek MDO3024, and 3 different 'headless' oscilloscopes) none of the headless ones are completely functional any more and all of the 'headed' ones are. But there is a saving throw here.

Like you, I appreciate how cost effective it is to build these things these days. Why not build the display as well? Lets say you contract with a tablet maker in Shenzen or build your own LCD display + SoC of your choice. Then you are in control of both the display and the instrument and it opens up some other possibilities. You could for example have several test instrument "bodied" that could pair with the display, so a DMM body, a scope "body" a Freq Counter body, etc. Easy to unbundle, easy to get either "higher end" or "lower" end remotes. Second it solves your multi-channel problem if you tie all the probes to the same 'sync' line. Now you can have 8 channels if you want and the display body software just sucks them all in. You can also dump Bluetooth LE and go with the inexpensive Nordic 2.4ghz spread spectrum radio chips. Now you can share a nanosecond disciplined time base with all your tools and bring the signals back together in the display "panel." You can sell a larger 'indoor/bench' display panel or a ruggedized 'on the road' panel. That iPad Pro in the video cost the person $700 - $800, a panel with an ARM53 SoC talking to it can be had in Shenzen for $35, $150 is you put an IPS panel on it. Now you sell it for half the price of an iPad, it works with all your tool heads forever and I, as the engineer buying your stuff, know that even if you go out of business my test equipment will still work.

If you build something like this I'll buy one for sure. Market size is somewhere between 1 and several million engineers :-). If you want more thoughts on this feel free to contact me in email in my profile.

j / k navigate · click thread line to collapse