A
Anonymous
Guest
.
A
Anonymous
Guest
.
A
Anonymous
Guest
.
A
Anonymous
Guest
.
I like some of those DIY units. But for now I'll have to stick to my trusty old...er, well, old, anyway...Tektronix 531A. 
A
Anonymous
Guest
.
Uh, I *might* be able to fit it on my *trailer*.nutsandvolts said:Can you fit it on your handlebar?
But I'd also probably burn out my old inverter trying to run it for any length of time, and it'd sure kill my traction pack! 
Yeah, I'd love to do that. I just don't have all the parts needed for any of the DIY scopes (except some of the soundcard scopes, but then I'd have to carry the laptop with me, too), and money is not available to buy stuff.C'mon I gotta have scope on handlebar.
That is interesting. Very interesting. Hmmm.
You keep finding all this stuff that is relevant to the ebike control system I wanted to make; some of it adds more ideas than I had to start with and some of it just makes the original idea easier to do/use or basically better. I wish one of us lived a little closer to the other.
A
Anonymous
Guest
.
I've been looking for a DIY Digital Storage 4-channel oscilloscope that has a bandwidth of 20 mhz or greater (So 100-200 Mega-samples/second), that also includes logic channels, and uses a computer for longterm data storage, visualization and other fancy stuff. I've so far have had no such luck in finding one. Most of the "DIY" ones are really cheap and that usually entails really crappy bandwidth due to A/D speed limits. The first one you posted to looks interesting, though, in that it seems the bandwidth could be expanded. I'm interested in the "DIY" aspect because then the software could be customized to my liking / requirements - it also seems like it might be a little cheaper than the existing commercial alternatives, which are priced at something like $1000+. There's USB scopes on eBay for $200, but I really don't trust the software would do what I'd want to do and I wouldn't know how to interface with the scope to make my own software for it.
I think it'd need a dedicated high speed A/D converter, and then that'd data would be sent directly to memory (Possibly RAM)? The memory would then be continually uploaded by a microcontroller interfacing with the computer over some connection. It seems, though, the amount of databits you could record in one continuous session would be limited by the memory as the data-sending speed looks like it'd be slower than the A/D converter speed unless there's some ultra-fast uploading possibility. It'd only need to upload at 15-20 Mb/s? I think ethernet is capable of that, but I wonder if the overhead would significantly impact the transmission speed or if I would need to implement extraordinarily complex transmission algorithms. I guess I could change the data recording rate depending on the needs so I could expand the amount of recording time available.
I wonder if you could use a single A/D for multiple channels via multiplexing. Are multiplexers fast enough for the task? For four channels, it'd only need to be able to switch somewhere south of 1 Gigasamples/sec, methinks.
I think it'd need a dedicated high speed A/D converter, and then that'd data would be sent directly to memory (Possibly RAM)? The memory would then be continually uploaded by a microcontroller interfacing with the computer over some connection. It seems, though, the amount of databits you could record in one continuous session would be limited by the memory as the data-sending speed looks like it'd be slower than the A/D converter speed unless there's some ultra-fast uploading possibility. It'd only need to upload at 15-20 Mb/s? I think ethernet is capable of that, but I wonder if the overhead would significantly impact the transmission speed or if I would need to implement extraordinarily complex transmission algorithms. I guess I could change the data recording rate depending on the needs so I could expand the amount of recording time available.
I wonder if you could use a single A/D for multiple channels via multiplexing. Are multiplexers fast enough for the task? For four channels, it'd only need to be able to switch somewhere south of 1 Gigasamples/sec, methinks.
A
Anonymous
Guest
.
Drunkskunk
100 GW
Having watched my professor fry a $2000 scope while playing with a tesla coil, i've been resistant to try ine of the PC based scopes.
That little $35 job looks like fun. It might not do everything I need, but I'd do all my dumb shyt with it before hooking up my computer to what ever insain invention is in my garage this time.
That little $35 job looks like fun. It might not do everything I need, but I'd do all my dumb shyt with it before hooking up my computer to what ever insain invention is in my garage this time.
A
Anonymous
Guest
.
I was more interested in borrowing the PC to mainly do the visualization and BIG data storage. It seems easier to graphically develop software on Windows then it does an embedded architecture, and just easier in general to manipulate data without worrying about "Blowing the stack". I also think such a "PC scope" would be the cheapest, since you don't have to develop the visual part with all the related display, chips, circuitry, etc.
A
Anonymous
Guest
.
Oh yeah, those look good. However, they're about as expensive if not more than expensive as commercial USB scopes. (Like the bitscope)
Looking at the chips on the board and the listed capabilities, I can guess that there's less than $100 in chips on the $1200 model, so I know a lot of the price is due to a low-volume niche market which really provides the DIY motivation.
(Like the bitscope)Looking at the chips on the board and the listed capabilities, I can guess that there's less than $100 in chips on the $1200 model, so I know a lot of the price is due to a low-volume niche market which really provides the DIY motivation.
A
Anonymous
Guest
.
A
Anonymous
Guest
.
A
Anonymous
Guest
.
nutsandvolts said:
Yes, very interesting. I wonder if it might be able to continuously stream the data to the computer so the "built in" memory isn't a time-limiting factor (for recording)? I also wonder how easy it might be to synchronize two scopes in the software so you could effectively build a 4-channel scope. Now I just have to find a PC logic analyzer (that's interface-able with custom software) to add to it. Heck, it'd seem pretty simple to build a logic analyzer to send the signals directly to the computer (comparators directly outputting to the parallel port with fast opto-isolated inputs?). I just don't how fast a simple design could be made.
Similar threads
