For about a year I have owned a Hantek 8060 dual channel oscilloscope. I have been very pleased with the scope as it is well suited to my needs. I use it in my simple experiments and I use it for helping me diagnose equipment that is in need of repair.
About three weeks ago I was working on a Denstply Cavitron which is a device used in dental clinics to clean scale from teeth. The Cavitron produces a 30 kHz vibration in a metal tip that quickly removes plaque and scale from teeth. The device operates on a magetostrictive design that couples a magnetic field produced in a coil with a core of metal laminations attached to the tip. I had completed repair of the device but I wanted to see if I could tune it more perfectly to the resonance frequency.
As I thought about how to hook the Hantek oscilloscope to the circuit to properly monitor the output I hit upon an idea that would prove to be ill advised and fatal to one channel of the oscilloscope. Rather than tap into the output of the circuit I decided to use an external coil to couple to the existing field generated by the Cavitron to produce the tip vibrations. Here is a picture of the coil that I chose to make this coupling.
You may recognize this coil as the primary winding from a conventional power transformer. The secondary windings and the metal core material have been removed. In my delight at having thought of an easy way to make my measurements I was no longer using the common sense that I should have developed in the last 60 years of working with electronics. I did not stop to think that the coil I was hooking to the input of my scope probably had upwards of 100 windings for each winding of the magnetostrictive cavitron handpiece. I did not stop to think how well the 30 kHz current in the handpiece would couple to the transformer coil. Quite frankly I did not think and when one does not think there is usually a price to be paid.
When all was hooked up and the Cavitron was turned on the channel one of my scope made a series of vertical off scale lines and then the trace stopped moving and settled permanently at the top of the display.
If you look closely you can see the yellow trace still pinned against the top of the display. Who knows how many hundreds of volts I had introduced into the input. I had blinded my trusty little friend in one eye.
( I would discover later that there was enough voltage produced to arc across and wreck the 1X 10X switch on the probe I was using as well as arcing across a couple of the legs on a SO-8 integrated circuit in the channel one input section of the oscilloscope. )
The first thought was to see if I could send the scope somewhere to get it fixed. After contacting several test equipment manufacturers around the US it became apparent that no one was interested in servicing an oscilloscope that only cost $400 to begin with. I wrote to the company where I bought the scope but they were strictly in the business of selling and did not have any idea of how to get it fixed. Finally I decided to write to Hantek itself to see if they had a way for me to ship it in for repair. I was skeptical of this as shipping to and from China does not give me a comfortable feeling and at least the US to China shipping would be substantial.
I was pleasantly surprised when the next day after my inquiry I received a reply from Amy Cai at Hantek. She asked if I was up to doing a little soldering and gave me information on the likely point of failure and the part number of the chip I would have to replace on the circuit board. She included pictures of the circuit board and highlighted the chip that was likely failed. Now working on Test equipment usually makes me quite nervous but in this case I really didn't have too much to loose so I decided to buy the IC that I needed, an SST5912 by Calogic. I looked on the Newark site, the Digi-Key site, the Allied site and every other electronics parts house that I had ever dealt with but none of them had any of the SST5912 ICs. I did find a couple places in Europe where I could buy quantities of 1000 for about $3000 but this was not going to happen. I tried EBay where one can usually find unusual parts and there was one vendor in China that still had 2 of the chips for sale. I had originally been worried about getting an aftermarket or knockoff part but this is probably quite unlikely as a knock off company would be stupid to put any production time into a part as obscure as this one. The part on Ebay was ordered and as of this writing is suppose to arrive in 4 days.
Since I decided to attempt this repair myself I have been stressing about the removal of the original chip from the circuit board. While I have done this many times and the chip is small, an SO-8 package, it is still within the manageable range for me. I do not know how many times I have mentally removed the old SST5912 from the board. I believe I have even removed it a couple times in my dreams. Unlike the time before the damage was done I have had my thinker running in full gear. Today in anticipation of the parts that will come I decided to finally open the scope and remove the old chip. Here are some pictures of the disassembly procedure:
These plastic covers hide the magic screws that hold the case together. The battery pack was removed prior to disassembly.
Here is the back of the scope with the cover removed. The metal housing in the center top is the input section of Channel One.
With the metal cover removed we can see the chip U30 that was identified by Amy Cai as the likely failure point and to my surprise one can also see the remnants of an arc discharge on the circuit board. When I first saw this I was quite concerned that it involved the 5R1 resistor in the middle of the flash pattern but testing revealed that the resistor itself was OK. More likely the arc came across two of the legs on the U30 and just contaminated the board with flash debris.
Now was the moment I had dreaded, removal of the old chip. I decided to use a technique that has worked for me successfully in the past. Using my best flush cutter I will snip each lead of the SST5912 and then after the body of the chip is removed I will sweep up the legs of the chip with a solder iron. The flush cutters do put some lateral force on the lead being cut so it is important not to produce so much that the solder pad itself is disturbed. The process of cutting the leads was done slowly and carefully.
Once the body of the chip was out of the way the legs of the chip were easily swept up by the solder tip and a final touch was put on the prep work by using the desoldering tool to suck the excess solder from the pads. I also took the opportunity to use a little flux remover to clean the arc flash from the circuit board. Here is the foot print of U30 ready for the arrival of the replacement chip.
At this point I will put this project on hold until the new chip arrives. I am not worried about installing the new chip on the board as this is a procedure that I have done many times and it is not that hard a procedure. I will begin to worry at this point if the damage was confined to the U30 SST5912 IC. I guess if this doesn't solve the problem I will have to write back to Amy at Hantek and ask her what the next link in the failure chain would be.
Update: The replacement SST5912 chips arrived today and I really hoped to be writing a success story tonight but it will have to wait. I have installed the new chip but the scope channel one still doesn't work. The pegged trace is no longer stuck to the top of the screen however. Now it is pegged to the bottom of the screen. Here is a picture of the replacement chip installed.
I have no schematics but I have been doing some reverse engineering and studying the other chips in the front end area and I think the chip I just replaced is an input protection circuit. The output or protected line from this IC leads directly to one of the multiplex inputs of a HEF4052. Perhaps that was also damaged by the jolt that I gave the poor oscilloscope. I have written to Amy Cai at Hantek in the hopes she will guide me to the next fallen domino in this circuit.
I have been debating whether to post this blog or wait for a more definite conclusion. I always hate it when I see the To Be Continued on the screen so I guess I will wait so you can celebrate the recovery or funeral without waiting for the next set of chips to come in.
Midnight and I have received a reply from Amy at Hantek. Her suggestion is to have the oscilloscope perform a self calibration. O K I think to my self, there is no way that it will be that simple but I am game to humor Amy. To my surprise the self calibration procedure puts Channel One back into operation. I am smiling ear to ear and I am testing and comparing channel one to channel two just to see if everything is working properly.
Now all I have to do is remember not to do anything as dumb as hooking an unloaded inductor to the input of any of my high impedance test equipment in the future.
John
Top Comments