FAQ, Tips, and Troubleshooting [Updated Oct. 11, 2017]

FAQ, Tips, and Troubleshooting [Updated Oct. 11, 2017]

Postby jye1 » Wed Feb 22, 2017 4:06 am

Frequently Asked Questions:

Q1: I couldn't calibrate C3 using the 0.1V test signal. What can I do?

A1: The current firmware has a bug that cause the 0.1V signal contains a large (3.3V) DC offset. When 0.1V test signal is enabled the trace will jump out of screen area. The workaround is setting the couple switch to AC position. This will block the DC offset and allow you to perform the calibration normally. Note: This bug has been fixed in firmware version -055.

Q2: The rotary encoder only do increasing no matter which direction it is turned towards. Is the encoder defective?

A2: This problem is likely caused by that the rotary encoder was installed on the wrong side of the small PCB on which the encoder is to be soldered. Please make sure rotary encoder is mounted on the side with encoder outline marking.

Q3: Why my voltage numbers are incorrect?

A3: There are two possible causes for wrong voltage display. 1 ) The reference point has not been calibrated, and 2 ) the gain of analog channel is incorrect. For 1 ) you just need to perform VPos alignment as instructed in the manual (it is strongly suggested to do factory default restore before VPos alignment to make sure all parameters are valid). For 2 ) please refer to the Troubleshooting section below to find out possible causes in analog channel.

Q4: How to update firmware?

A4: Please refer to http://www.jyetech.com/Products/LcdScope/DSO150_HowToUpgradeFirmware.pdf. Pay close attention to the signal line connections. The transmitting line of scope (TX) should be connected to receiving line of USB-Uart converter. The receiving line of scope (RX) should be connected to transmitting line of USB-Uart converter.



Assembly Tips

1: Calibrating C5 and C3

From time to time we received user complaints about that turning C5 did not have any effects. One cause of this issue could be defective C5. But more likely it is because of that the ratio of parallel value of C5 and C6 (designated as C5//C6 below) to C4 can not meet the requirement (which is about 100:1). To understand this let's delve a little bit into theory.

In the circuit R3 and R4 define the attenuation ratio we need (about 100:1). Because of the existence of U1B input capacitance (Ci) and stray capacitance Cs R3 and Cs//Ci form a low-pass filter (see Fig. 1) which makes high frequency signals are attenuated more than low frequency signals do. To overcome this problem we add C4 in parallel with R3 to allow high frequency signals to pass through easier. What we want is all signals be equally attenuated. For this purpose the value of C4 can not be arbitrary. It can be proved that to achieve a flat attenuation the ratio of C4 to Ci//Cs need to be the reciprocal of the ratio of R3 to R4. That's to say the ratio of C4 to Ci//Cs needs to be about 1:100. Users can google the net to search for the mathematical reasoning behind.

CalabratingC5_1.jpg
Fig. 1


The value of OpAmp input capacitance Ci ranges from a few pF to tens of pF. The stray capacitance is usually less than 1pF. So C4 needs to be very small. Ci is different from chip to chip so C4 needs to be different too. It is very impractical for this application to use a such small variable capacitor since it would be expensive or complicate. So what we did was making C4 fixed and adding trimmer C5 and fixed capacitor C6 to parallel with Ci and Cs in hope that by adjusting C5 the ratio C4 to C5//C6//Ci//Cs meets our requirement (see Fig. 2). C5//C6 is significantly larger than Ci//Cs. A benefit of so doing is that it plays down the affect of the Ci/Cs change caused by change in temperature and humidity. In the circuit C4 uses 1pF. C5 uses a 6 - 25pF trimmer. C6 uses 100pF (or 120pF).

CalabratingC5_2.jpg
Fig. 2


The theory above sounds good. But in real world there are some issues in this circuit. Firstly, C4 has errors. For common ceramic capacitors the tolerance of C4 is about 10 - 20% unless expensive capacitor is used. Secondly, the variable range of C5 is small comparing to what we need. That means the coverage of C5//C6 may not be able to achieve the ratio we want if the value of C4 stays away from its supposed value too much. Again, it is impractical to use expensive wide range variable capacitor here for this low cost product. This is the problem we have and the reason of many users found C5 not functioning as it should be.

One solution to the problem is changing C6. If we know C4 is too large or too small we can replace C6 with a larger or smaller capacitor so that C5//C6 can cover the wanted value. By looking at the waveform of test signal you can tell which way C4 goes. If you see overshoots C4 is too large. You need to make C6 larger. Otherwise, if you see round angled waveform C4 is too small. You need to make C6 smaller. You can also replace C4 to achieve the same goal but pF range caps are not as easy to find as bigger ones.

There is another solution that could be better and easier particularly for users who do not have spare caps floating around. In this solution we use a home-made trimmer to replace C4. Get a piece of wrapping wire (AWG28 - AWG32 size is good) and cut it to two segments each in 3cm length . Twist them together and strip and tin one end as shown in the photo below (Fig. 3). A home-make trimmer is created. Solder the tinned ends to the place of C4. Now if you want the capacitance of C4 larger just twist the wires tighter. If you want it smaller you simply cut it shorter (cut little by little to avoid cut too much unintentionally). Don't forget you can also adjust C5. In this way it would be much easier to obtain the C4 to C5//C6 ratio we need. After finish use some glue or silicone to secure the trimmer so as it keeps the capacitance set.

EasyCapTrimmer_2.jpg
Fig. 3


EasyCapTrimmer_3.jpg
Fig. 4


For C3 calibration similar method can be applied to C2.


Troubleshooting

1: Test signal did not show.

1 ) First verify all voltages at the check points on the analog board are correct and make sure there is a trace on the screen.

2 ) Make sure your scope can react to input signal Set the couple switch to position DC or AC. Touch the red clip with your finger and see if you have reaction on the screen. If you don't see reaction touch the center conduction of BNC connector with a tweezers. If you still can't see reaction carefully check the soldering of analog board and verify the voltage at check points. If you see reaction at touching the BNC center core that indicates the probe might be open. Check the continuity of the probe.

3 ) Make sure test signal is available Set test signal amplitude to 3.3V (the default setting) and use a DC voltage meter to measure the voltage of test signal (the negative pen of your meter should be connected to ground). It should read about 1.6V. If you get 0V you need to trace up to see where it was disconnected. The test signal comes from pin 17 of U1 and connected to J8 through R16. First measure voltage at R16 both ends. If you don't see measure again at pin 17 of U1 (you need a sharp object to measure voltage at U1 pins. You can use a segment of component lead and manage to connect it to your meter pen). By this way you should be able to uncover the problem spot if there is open.

2: No negative power supply. V- and AV- have a positive value less than 1V.

1 ) First check the voltage at U5 pin8 to make sure power supply is connected to the chip. This voltage should be the same as V+.

2 ) Visually check all pins of U5 to see whether all of them are soldered good. Check the soldering of C12 and C13 as well.

3 ) Check C12 and C13. First use an ohm-meter to check whether they are shorted inside. If they are shorted replace them. If they are not shorted then connect in parallel with a similar value capacitor and see if you can get V-. If you do replace the old capacitor.

4 ) Disconnect one end of R27 and see if the you can get negative voltage at V-. With R27 being open the loading is cut. This is to eliminate the possible impact from loading.

5 ) If V- is still missing with R27 being disconnected it should be considered to replace U5.

3. Incorrect Analog Gain

Understand the circuit

Signals is input from J1 and fed the front stage through couple selecting switch SW1B. There are two branches for the front stage. One is the high gain branch which consists of R1, R2, and U1A. The other is the low gain branch which consists of R3, R4, and U1B (at this branch signals are attenuated to about 1/100 of its original value). SENSEL3 selects which branch is connected to the middle stage (U4C). When SENSEL3 is HIGH (which corresponds to sensitivity from 5mV/div to 0.2V/div) the high gain channel (U1A) is selected. When SENSEL3 is LOW (which corresponds to sensitivity from 0.5V/div to 20V/div) the low gain channel (U1B) is selected. The middle stage amplifies signals by 2 and feeds them to the main attenuation network formed by U3 and R7 - R12. Signals SENSEL0 - SENSEL2 determine the attenuation of the network by selecting different positions of analog switch U3 (as result, different sensitivities of the analog channel). The last stage (U4B) is the main amplifier and drives ADC through R16. Resistors R13, R14, and R15 set the gain of U4B to about 8. U4D, D1, R21, and R22 set the level for 0V input. This level is about 1.5V. U4A, C8, and R17 - R20 form the hardware trigger. Please note that for firmware version 113-15001-060 or later this trigger is not used due to improvement of internal algorithm.

Please follow these steps to check.

1 ) Check and make sure all resistors are of correct values and installed good. R1 - R15 are gain setting resistors so pay close attention to these resistors. Remember, if you measure resistors in circuit the results usually appear smaller than their real values due to measurements are affected by other in-circuit components.

2 ) Check the signals SENSEL0 - SENSEL3 by referring to the chart below (Fig. 5). These signals determine the gain of analog channel.

SENSEL[3-0]_Levels.jpg
Fig. 5 SENSEL[3:0] Levels vs sensitivity settings


3 ) Do not forget to check power supply. If AV+ and AV- are too low the working range of amplifiers is reduced which could cause gain error. In most cases AV+ > 4V and AV- < -3.5V are considered acceptable.


4. Incorrect Display of Couple Setting

This issue is often related to the parts SW1A, R23, R24, and R25. These components form the detection circuit for couple switch (SW1B) position. The MCU detects SW1 position by measuring the voltage at its CPLSEL pin (PA5, pin15 of U1). If you find the display is wrong check these components for correct values and good soldering. Please also check the voltage VBIAS which should be around 3V. This voltage could cause problem if it is off 3V too much.


5. Rotary encoder does not work properly

Most of time problems of the rotary encoder are caused by issues in its connections. The schematic below shows how the rotary encoder connects to MCU. Note that SW6 (which is not provided with the kit) in the schematic is not same the encoder as provided with the kit. The encoder with the kit is installed on a small PCB (PN: 109-15002-00A) and connects to MCU through pin-header J2. But both encoders share the same pin-to-pin connection relationship with regard to the MCU. Fig. 7 shows a board with rotary encoder installed. The pin pads to be used for checking are labelled.

EncoderConnections.jpg
Fig. 6
EncoderConnections.jpg (11.16 KiB) Viewed 3219 times



EncoderCheck_2.jpg
Fig. 7



Please following steps below to check the connections.

1 ) Check to make sure no opens between the following pin pairs. The resistance between these pin pairs should read about 1K.
DB0 ---- to ---- encoder pin 1
DB1 ---- to ---- encoder pin 3
DB3 ---- to ---- encoder pin 4

2 ) Check to make sure no shorts between ground and pin 1, 3, and 4 of the encoder. Resistance between these pins to ground should be greater than 10K. NOTE: At certain positions of the encoder pin 1 & 3 are connected to pin 2 (which is ground) internally. If you find both pin 1 & 3 read zero resistance to ground you just need to turn the encoder by one position to make the internal connections open.

3 ) Check the pull-up resistors R23, R24, and R25. Measure the resistance between pin 1, 3, & 4 and +3.3V respectively. The result should read about 10K for each pins.
Last edited by jye1 on Mon Oct 30, 2017 11:05 pm, edited 16 times in total.
jye1
 
Posts: 995
Joined: Wed Feb 20, 2013 4:58 pm

Re: FAQ, Tips, and Troubleshooting [Updated Feb. 2, 2017]

Postby tin » Mon Mar 13, 2017 4:25 am

Dang it... I feel so silly having the rotary encoder in the wrong way now. This is going to be fun to fix....
tin
 
Posts: 4
Joined: Sat Jan 07, 2017 4:40 am

Re: FAQ, Tips, and Troubleshooting [Updated March 18, 2017]

Postby nunoit » Mon Apr 10, 2017 11:49 am

i ended up in adding 100 pf in parallel to c6 to remove spike, c5 remains useless i tested all caps before installation.
next would be to install the software upgrade and if the test signal was less noisy,

for those that want to make c4 cat5/6 is 28 guage
nunoit
 
Posts: 14
Joined: Sat Apr 08, 2017 3:33 am

Re: FAQ, Tips, and Troubleshooting [Updated May 22, 2017]

Postby surenz » Wed Jun 28, 2017 6:51 am

I would suggest two additions to the firmware update guide:

1. I don't see why we need external power supply when most of TTL-USB bridges have nice 3.3V which can be used as power for the CPU during the programming. I've used it twice already and no issues so far.
2. One very important screenshot is omitted from the guide during the programming of the chip:

Kiss_flash_17.jpg

When programming for first time the CPU you need to click on 'Remove protection' button. Nothing major will happen as is written with red characters.
I spent some time researching what will happen when I click on this button and fortunately found a video in Youtube where someone programmed DSO138 with same software.

Next programmings don't have this button activated so you can follow the guide as is.
surenz
 
Posts: 2
Joined: Wed Jun 21, 2017 8:23 am

Re: FAQ, Tips, and Troubleshooting [Updated May 22, 2017]

Postby jan » Thu Aug 03, 2017 4:56 pm

will it be possible to share some wavefrom pictures , to get the interpretation of c4 //c6 right
jan
 
Posts: 26
Joined: Sun Jun 25, 2017 8:38 am
Location: the netherlands

Re: FAQ, Tips, and Troubleshooting [Updated May 22, 2017]

Postby jye1 » Sat Aug 05, 2017 2:43 pm

By waveform edges being overshoot or round-angled you will know C4 is too large (C6 too small) or the opposite.
jye1
 
Posts: 995
Joined: Wed Feb 20, 2013 4:58 pm

Everything running EXCEPT...V/DIV not working

Postby albenbrewder » Mon Oct 23, 2017 8:17 pm

This has been re-posted to 'Missing Button Response' section...



Ref: boards: 109-15000-00E (Display); 109-15001-00E (Analog)
Software: 113-15001-064
Initial Display board check showed everything working EXCEPT V/Div push button did nothing.
I have checked all connections and DB4-DB7 appear OK, (Pull downs, V->0 when pressed)
Board soldering appears OK... Problem appears to be either a cpu malfunction or a program bug.
Rest of project was completed and all appears functional EXCEPT I can't change V/Div.

Hardware?
Software?
albenbrewder
 
Posts: 9
Joined: Sat Apr 25, 2015 8:06 pm


Return to DSO Shell

Who is online

Users browsing this forum: No registered users and 2 guests