DIY Oscilloscope for Sensor probing?

Status
Not open for further replies.
Joined
Apr 9, 2008
Messages
16,461
Location
Central NY
So I have been watching a lot on youtube with people doing electrical diagnoses on vehicles. There have been a few cases where I have had a check engine light for something or another wrong, and I just buy a part and put it on and hope it works.

One of the features of the fancy $10,000 scanners is the oscilloscopes they have in them. Obviously they have 4,5 channels and these just have one.

It seems that these would work for probing , say, a crankshaft sensor on a Jeep. And would give a definitive answer as to why something stopped working.



DIY Oscilloscope

Anyone ever use one of these? Or think it would work?
 
I see they have come down in price--I was tempted when they were $100. $23 is cheap enough to try out. looks like they may have done the fine pitched SMT chips already; soldering thro-hole is a breeze.

That said, I'd have low expectations. For doing O2 sensors, not a problem; not sure what the higher PWM speeds are for sensors (several kHz). I suspect the input filtering isn't up to par and that aliasing will be a bear on it (then again I got bit by that the other day! on a good LeCroy no less) if you're not careful.
 
Originally Posted By: Virtus_Probi
You can also get one assembled for only $30, there is a link on the page for the kit.


LOL, this guy will solder it up for you for $7.



lol.gif


I have seen automated equipment to do through-hole, so I guess it's a plausible price bump, for overseas.
 
Originally Posted By: Donald
Learning how to properly use a scope is 10x the
learning time of how to use a DVM.

Back in the day Tektronix was king. Maybe still is.
n There are a lot of really fine Tex CRT scopes on the used market. I have a 100 MHZ Tex 465 I got for a hundred bucks, with a good probe.
 
I actually have a few of those! But they aren't terribly portable. And I wouldn't want to leave it in my unheated garage all the time.

I figure for $30 it wouldn't be too much of a waste and still a cool tool.

Originally Posted By: Donald
With a clamp-on amp probe you can look at the current draw of the starter and see basically the draw per cylinder and can detect a low compression cylinder.


I have been wanting to try that one the Jeep. It definitely sounds like it has low compression on one cylinder.
 
They might have some good DVM's today but debugging a crank sensor isn't going to be something to do on a low end DVM. Depending upon the application, "wide" and "narrow" pulses are to be expected: some OEM's (if not all) use a signature tooth so as to know exactly where they are on the target. One tooth may be 2x the size of other teeth, thus telling the ECU TDC. As a result, a graphical display may be required.

Only real gotcha is if the sensor has an output which is a current signal, instead of a classical 0 to 5V swing. Even still, with a good DSO, you should be able to see that, albeit with a smaller voltage swing (half volt or so) and usually riding on top of VCC, unless if the OEM went real fancy in reading that sensor (unlikely).
 
Besides only being one channel, about the only thing I see wrong with it is that the peak to peak voltage is only 50v. When reading injectors or coils, the collapsing fields can spike well over 100v, so you would need attenuators to knock the signal down- and then do the math to figure out what you're seeing on the screen. You would also likely need more adapters and widgets for additional automotive testing.

For a (relatively) cheap auto style scope with a fair assortment of accessories, look up AES Wave's Uscope.


edit: I've got less than 10 grand in my equipment, but only by purchasing wisely. Sure you could spend 10k or more on the fancy Snap On Verus, but really there are plenty of other scan tools that will do the same scan tool funtions as the Verus for WAY less. Then you take some of the money and buy a quality lab scope. I have the Snap On Solus Edge and the Pico 4425 lab scope (which is really a better scope than what the Verus comes loaded with) and still came in well under the cost of the Verus.
 
Last edited:
Looking at it again, the nice thing about it is that it uses a BNC input, all the common accessories should plug right in.
 
I really want to get a Picoscope. Not cheap, but not bad either. USB powered. Just use my laptop to power, and screenshot, issues. Saving to disk is nice (any scope I buy *must* have that, kinda worthless otherwise). Have a 4224 in my office, only 20MHz BW but it suffices for simple tasks.

Personally, I wouldn't mess with anything less than 10MHz and 4 channels, but I suspect most auto debug would be ok with 10MHz BW and two channels, unless if digging into CAN buss or the like.
 
that's a pretty cool little kit. probably fine for auto use - note the statement about attenuation above - agree.
 
I have one and it works ok. But, the display isn't very bright and it is hard to see in bright lighting. Also the text etc is pretty small on that size screen.
It was easy to build and it works, but for me it is more of a toy or maybe a dedicated display on another project.

Rick
 
Miller88- thanks for the link, I ended up buying one for my cousin who dabbles in auto work on the side. I figure it would be a good learning
tool and if he blows it up then nobody is really out anything. Once he gets proficient, he can buy his own, more capable scope.
 
If you have a Raspberry Pi(it's a small credit card sized computer with 2 USB ports, HDMI out, Ethernet and GPIO headers that has the computing power[roughly] of an iPhone 4S), there are kits to make an oscilloscope out of them.

http://www.bitscope.com/pi/
 
Last edited:
Originally Posted By: Vikas
The 1 MOhm impedence is not good at all for that $30 scope


Last I knew, 1Meg was pretty standard for scopes. 10Meg is from using a 10x probe.
 
Status
Not open for further replies.
Back
Top