I am currently building the Video 43 version of the capacitor tester. I’m wondering if anyone has had the same experience as me. Upon first power up at 24.5Vdc, I noticed a few of the LEDs were not illuminating as I adjusted the display driver adjust VR. I gave the board another look under the microscope and found a few of the LED driver pins were not completely soldered. I soldered those pins again to correct that issue, but now the last four pins of my display are not being driven from the LM3914V. I should clarify that those LEDs were included in the original problem mentioned above. I am worried that I may have possibly overheated the LM3914V. I have checked and re-checked for proper component placement, solder issues, and trace problems. Also, I can illuminate all of the LEDs with my bench multimeter diode check functionality. With no power applied, I hold the positive lead at the LED common rail and run through the cathodes of each one individually with the negative lead.
If anyone has had a similar problem, please let me know. I am not really looking for an answer to my particular problem, but rather guidance from those who may have experienced something like what I am seeing. I am confident the problem must be something simple I am missing. Thanks in advance for any help!
First of all, let me assure you that the device works great and the project is repeatable no problem at all. Now, to your problem: I would make sure that all values are being placed in the right place (mixing up between components can easily happen). The IC itself (as being an old design) is usually quite resilient to burning by overheating (in my experience). Depending on your SMD soldering skills but it could be that some IC pins are still not soldered even though they might look to be.. Make sure that the IC has been soldered the right way around (check the IC’s key marking). Another problem could be with either low-quality or defective VR’s. I would check those with a resistance meter to see if the cursor on each of them runs smoothly (with no sudden jumps) throughout. Triple-check everything. Test the voltage across the zenner to make sure it regulates the voltage at or very close to the voltage stated on the zenner itself. Beware that underneath the 100K VR there is a fine trace. Make sure it does not short (touch) with the VR itself. This is a known problem. If needed, fit a small paper piece under the VR (to act as a spacer) before soldering it in place. After that (with a pair of tweezers) pull the paper piece out. Verify that you didn’t “Somehow” solder an NPN in place of a PNP transistor and the other way around (it can happen as well). The Component Layout Map PDF document is your best friend! Make sure that you strictly follow the calibration instructions. (Re-watch the video as many times as needed to understand it. Pay special attention to the “switch position” you need to be in during the calibration procedure. (See video). Another trick is to have everything “super clean” and no flux residue anywhere to be seen on the PCB. (Very important to do it before attempting any calibration procedure.) Make sure that you only power up the tester using a proper Linear PSU as opposed to a switch mode power supply. SMPS (generally speaking) are known to be “Noise generators” and they could impede the calibration and/or functionality of the device! I hope it helps.
PS: pay attention to the way you solder the wires to the SW. (it can be tricky!) and that the switch itself is the right kind. It has to be of the “On-On-On” kind. But then again, if you’ve purchased the exact part number specified in the “Parts List (PDF) you’re safe. Now regarding the high Ohm value resistors, they are crucial to your success! Also regarding the 2 high-gain (2N…) transistors make sure you have the means of checking the “Beta” value for each of them and that the measured value is within the “ballpark” of what the datasheet specifies (very important!)
Thanks for the reply! When I have time to work on the project this weekend, I will follow your suggestions and let you know what I find.
@ovi4 That is one of the PCB I made, but have not started to work on it.
I am looking for any problems I may have making it.
Larry – N7LUF
Update! It has been some time since I last visited this project. Work and life always seem to get in the way of our hobbies. I am pleased to say that I have resolved the problems I was having with this circuit. As Ovi4 mentioned, the circuit is reliable and reproducible. I found that the LM3914 that I originally soldered onto the board was damaged either from the factory or from me mishandling it (possibly ESD?) I damaged my original etched PCB while attempting to remove the non-functional LM3914 without a hot air rework setup ( I have since resolved this deficiency). I then fabricated a new PCB and populated it with all of the passive components from the original board. I was cautious with the active components because of the potential of thermal damage during the soldering process. Long story short, I now have a perfectly functional Ultra Sensitive Capacitive Leakage tester.
In my case the original LM3914 was the issue. These things happen sometimes. Thank you Paul for all of your hard work and for sharing your designs with the community!
@ingrownelectron I have just got thinking about ESD as I am new to SMD parts and how I should protect what I am doing.
I got a message that from Mr. Carlson he has a video in the Q for a video on ESD.
I wonder if there is a away to test a part to see if it is damaged, or it works or not.
I looked and I have 10 LM3914 in my supply and I need to know more about ESD.
Larry – N7LUF
@ingrownelectron I’m glad you’ve solved the mystery issue. Now regarding the IC. : this particular IC is kind of hard to solder since is not an SMD style part and it “usually” requires a specialized socket for it. The problem is that the vast majority of people (observing the past Patron comments for the project) are having is regarding the difficulty of properly soldering of the IC. since the pins are bend around the ICs body and they end underneath. Another issue is adding way too much solder to it and some hidden bridges would form (right underneath the IC) and one wouldn’t be able to see it unless you completely un-solder the entire IC and start again. The same issue might have happen to you as well and you didn’t even noticed it. The ICs in question are “usually” fairly resilient to temperature abuse therefore. I’ll be surprised if that IC is really damaged. However, there is a small chance. Now, regarding the hot air tool: it is actually a very useful addition to your tolls “arsenal” if you need to unsolder things (and especially ICs) that are SMD style mounted. It would save you the struggle and most importantly the trace damage.
@larry_n7luf testing an IC (generally speaking) regardless of what kind they are “usually” requires testing rigs purposely build for such ICs. When MCL came up for the very “first time” with the above IC idea (that was back in 2018 in videos 40-41) and shortly after that I decided to built that version of the project but was having all sorts of weird issues with the project (not necessarily the IC) but was however thinking of a testing rig IDEA for this IC family. 3914.3915, 3916) because I’ve predicted that the IC will also be employed on some other future projects. Something that has “Arduino style connectors” so you can plug in 2 different IC packages via 2 individual “IC breakout boards” one for the DIP18 socket and another for the PLCC20 socket . However as the time went by and being busy with all sorts of other thing the idea has been left “up in the air” ever since. 😏
@ovi4 I should have mentioned in my follow up post that I did actually test the original IC. Once I had the PLCC package extracted, I cleaned the pins and tacked kynar wire to each one. I then tacked those wires to another test board that I designed on perf-board specifically for this reason. The last four LED driver pins were not functional. *Shrug* I agree that these ICs are generally very robust. Also, the circuit that MCL designed is easy enough to understand. I was just “lucky” enough that my IC was damaged. Thank you all again for your interest and support. I’m happy to have access to a community like this.
@ingrownelectron Yes. a stand alone testing rig is indeed the way to go, to properly test those ICs. Great stuff. Tnx.