• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

DSO138 Review/Notes/Improvements

Status
Not open for further replies.

MrAl

Well-Known Member
Most Helpful Member
Hello there,

Recently i had obtained one of these little scopes and read some of this thread:
http://www.electro-tech-online.com/threads/diy-audio-scope-cheap.144010/

That's an older thread now so i am starting a new one here.

First i should note that i intend to be very critical of this scope in order to bring out all the details, but i want to make it clear that the scope is usable and is still quite amazing for it's size and price. There are drawbacks, some of them coming from the host hardware chosen for the device and some of them from the design which is very sub optimal. So in general it is a good scope but could be improved without affecting the cost of the device.

First the hardware. The core of the scope is the ARM chip, which has a 12 bit ADC onboard and functions as the real heart of the scope. It looks like not all 12 bits are being used however as the input to the ADC is limited more than it needs to be, and for offsets that are 1/2 screen we loose 1 bit right off.
The op amps are TL084 which although are great at plus and minus 15 volts, might be questionable at plus and minus 5 volts. We could check this better to find out what is happening.

The biggest problem seems to be with the front end, which appears to be poorly designed and which, if improved, would quickly clear up some noise problems on all but the lowest input range just like other scopes. One design flaw is that this front end is set up to gear everything down to 10mv per division and then amplify anything needed back up to the normal ADC input range. That's not the way to do it properly.

The other thing that limits the resolution is the resolution of the screen itself. Since it is 320 x 240, we can only see at most 8 bits on screen. This is hard to get around because a larger screen would be more expensive, but a vertical scroll would help this quite a bit so that we could zoom in on the wave and get the full resolution available with the onboard ADC, providing we also improve the front end.

The lack of a variable manual horizontal sweep setting. This allows us to view entire waves on the screen even when they dont fit on any preset horizontal sweep setting like 1ms, 2ms, 5ms, etc. There is a way around this by scrolling left and right, which is very handy, but scrolling the whole wave takes a good 30 seconds of holding the button down. It would be great to see this sped up in order to scroll faster.

What i dont know yet is how and when they are storing various values in EEPROM. I will try to find this out. What i worry about is that they store every single scroll event in EEPROM, which would wear it out much faster than if they just store some, or none, in EEPROM but rather in RAM. It doesnt matter that much if the scroll does not come up the same on power up. They also store other values like the verical setting, the horizontal setting, and that might be happening during use. If these are only stored when power down, that would be good, but i dont see them doing that as there are no large caps to keep power alive during power down. This means EEPROM could wear out too soon if the device is used every day. One the EEPROM wears out it may be impossible to set the device as needed.
This is an area which is hard to evaluate though until i find out more.

I have used mine several times now for viewing a few different things like an IR remote control signal using a photo diode as detector. The waveform is reasonable and it makes it easier to see the wave due to the storage function of this scope. My bigger CRT scope does not have a storage feature so viewing a wave that happens only once when you press the button is much harder on that scope, and when new that scope could have been easily 500 dollars (USD) although used it's probably not worth anything now that it is so old :)

The other problem that appears to be in the front end is the bandwidth. The bandwidth i measured on the 1v scale is about 120kHz not 200kHz, but i did not check the other input ranges yet. To redesign the front end we could just create a new front end and disconnect the old one and use the new one. That's if you want better performance and you dont mind doing a little work. We could also add a range doubler, as well as an input amplifier to get down to 1mv per division for example. In fact, an external input amplifier would do that too without modifying anything else.

One thing i have not decided yet is what kind of case to use with this. I see them sold on the web, but they are made of acrylic plastic not ABS. Also, i wanted to build mine in a case with a battery box to make the entire scope completely portable for measurements where power is not always available, such as in the automobile.

I also might get into improving the design a little to get it up to 100MSPS which would vastly improve the type of waves that can be viewed as it is just 1MSPS as it comes. That in conjunction with a new front end of course. This i would consider a more major upgrade however.

I'll add more if i find out more about this scope and anything related. If you have any information you would like to add that would be good too.
 

Ian Rogers

User Extraordinaire
Forum Supporter
Most Helpful Member
I bought a DSO201 which is pretty much the same... It was only £40 ish... I use it in the field.... Very smart... works very well.. Not a scratch on a proper DSO though.... The plastic design seems to be rugged enough and the slow settings ie... 100ms and slower seem to flicker like hell... Other than that... Does what I bought it for..
 

MrAl

Well-Known Member
Most Helpful Member
Hi Ian,

Yeah that looks more compact and easy to carry around. The DSO138 is just the circuit board so i have to carry around the battery pack i made separately, or else put everything into a box.
Looks like you also get 2 hours run time. Not sure how long i would get with 6 AA batteries yet. Using a 9v battery i get about 2 minutes before the battery runs down and wont power the analog circuits properly anymore.
Still thinking about a long term solution for the case and type of battery.
 
Status
Not open for further replies.

Latest threads

EE World Online Articles

Loading
Top