Fixing the JYE Tech DSO138 Oscilloscope

Click to enlargeAfternoon all!

I need to lift the 320×240 LCD display away from the main PCB, but have it still connected, so that I can fault find the negative supply for the analogue circuitry.

I came up with the idea of using an old IDE cable which has two sockets each with 2 rows of 20 pins on a length of ribbon cable. Unfortunately, the sockets have a missing pin presumably to prevent connection the wrong way round. Fortunately, the missing pin is not required by the LCD (as long as you connect it the right way round). So I’m drilling a small hole in the plastic.

Hmm, the IDE cable idea doesn’t work :(

It doesn’t matter how I orientate the ribbon cable, the pins don’t line up correctly.

IMG_20150612_151902[1]So on to plan B. I’m using a length of double row, long pin headers and lots of short female-to-female dupont wires to connect up all the relevant connections between the LCD and the main PCB.

Double check everything and apply power. The CPU seems happy (a double flash of the green LED). The LCD is still blank though.

IMG_20150612_152724[1]VGEN is present, a square wave of about 20% mark/space at 17.58kHz as indicated by my OWON oscilloscope. AV- is still 0.6 volts.

NPN transistor Q2 (9014) is looking suspicious – there’s a pulse train going into it’s base, but nothing on it’s collector.

Going to replace it with a 2N3904 which has the same pin arrangement. Part replaced, but no improvement.

IMG_20150612_161218[1]Some people have said that the inductor L2 has been a problem. Dig out my component tester and give it a test.

The component tester says it’s fine – 1.06mH with a resistance of 4.5 ohms.

Next, I’m not liking Q1, the PNP 8550 transistor. But it seems it’s OK too.

UNBELIEVABLE!! – These stupid 4-band resistors have ambiguous values. 120 ohms (brown, red, black, black, brown) is the reverse of 10k (brown, black, black, red, brown). I’ve swapped R12 with R37 and also R13 with R39. Now I REALLY HATE 4-band resistors :(

So with the resistors in their proper places, I now have -5 volts on the AV- test point. I also have a 2N3904 in place of the 9014 and a 2N3906 in place of the 8550, both appear to be good substitutions.

The user guide does suggest that all resistors are measured before putting them into the circuit, but it doesn’t mention that 120 ohms and 10k are the SAME COLOUR CODE!! Ridiculous.

IMG_20150612_171657[1]So, I now have all the proper voltages on the AV-, AV+, +3.3V and +5V test points, but the display is STILL BLANK!

Back to Banggood to ask for a replacement LCD.



4 thoughts on “Fixing the JYE Tech DSO138 Oscilloscope

  1. Thanks, Julian. That was very informative and helpful.
    I built this just for fun and experience so not a biggy that it’s not working.
    I have a working LCD but no display (backlight only)
    Also I have about +0.7v at the AV- test point. All other voltages OK.
    I don’t have a lot of soldering experience, so it’s possible that I fried the transistors, so I’ve ordered some replacements, (I’ll use a heatsink clip next time) but first I’ll try connecting the LCD to the mainboard with Dupont M-F leads ( not F-F as you stated, sorry, I know it was just a mistake -;)
    I don’t have enough leads currently, so I’ve ordered some more of those as well.
    Fingers crossed, I can sort this out.

  2. I was wondering about this oscilloscope the other day, and if you had received a replacement from Banggood. Will check back later to see if are going to post an
    update. All the best !

  3. Great follow up, Banggood has a multimeter kit you might want to have a look at.
    Anyways always enjoy your video’s keep them coming, Cheers.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>