To begin with, it was pretty crusty on the outside. At the top right is some glue-like stuff that was incredibly difficult to remove without damaging the face. It was all over the knobs as well. A couple of the push-buttons at the top had black marker dots on them. On top were pieces of foam tape and calibration stickers that had to be removed, followed by adhesive removal. The clean-up part of this task took about an hour. (Click any image to enlarge)
After cleaning it up I had to replace the cooling fan. The fan bearings were going out, and it sounded like a cement mixer when it powered up. As you can see in the below image, the face has cleaned up nicely. But it sounded awful. Really bad.
So I decided to perform some surgery and see if I could replace the fan. This board (A16 Processor Board) had to come off. Six screws and six ribbon cables, and then I could remove the board and see the cooling fan.
I had tried to oil the fan earlier and sealed the bearing with painter's tape, but it still sounded like a dying cow.
Below: The offending fan, after I got it out onto the workbench. The rotating assembly was incredibly wobbly. I could rock it side to side on the fan blades, the bearing was so worn out.
A few days later, this arrived in the mail. Three wires instead of two. The yellow wire is to adjust fan speed, which this oscilloscope is not set up for. No worries - I can cut that wire short.
However... The fans are not physically the same exact size. The one I needed was no longer available. Scope was last manufactured in the mid 1990's, and probably the fan as well. The new fan is just a bit smaller. But it won't fit without some modifications. I decided to make an adapter plate.
I'd kept a piece of duct work from last summer's furnace replacement, just for something like this.
I'm not a skilled sheet-metal worker, trust me on that! I have the cuts to prove it.
More or less cut to the dimensions of the old larger fan.
I drilled the first couple of holes, then put screws in them to keep the plate from wiggling, then drilled the last two. I didn't want the holes in the sheet metal to be misaligned.
Next, I set the new, smaller fan onto the sheet metal. I drilled holes for the mounting screws, and marked a circle where the smaller fan diameter needed to be cut out.
A bit rough-looking, but the fan should now fit mechanically...
After I installed the adapter plate and the new fan, it was of course slightly off-center, since I used one of the old fan holes to mount the new fan. I could have centered it, but it would have been more trouble and would not have improved the air flow anyway.
To improve airflow with the smaller fan, I cut out the reinforcing metal strips.
Above: It appears that the only reason this oscilloscope needs a cooling fan are the two semiconductors on the aluminum heat sink right in front of the fan. I'm pretty sure the bottom one is a thyristor in the voltage preregulator, and the top one is a voltage regulating zener diode in the final 51 volt DC regulating section. It shunts excess power to ground, so it gets hot.
Below, I soldered the old connector to the new fan, and cut the unnecessary yellow wire short, then put shrink tubing on the conductors to insulate them.
Below, fan power leads threaded through the panel and connected to the power supply board.
Moment of truth. Will it work properly after all this digging and scratching? Apparently, yes.
There was still an issue. The fan did not work. I took everything apart again and tested the fan on my benchtop power supply. It worked perfectly and drew the correct amount of current (0.37 amps at 12.0 volts. I put it back together again and it still didn't run.
Eventually I figured out that one of the mounting screws for the A16 Processor board sat right above the fan. The screw threaded down far enough to interfere with the rotation of the fan. I chopped a quarter inch off the screw and the fan ran fine. I hadn't even considered mechanical interference!
The only fly in the ointment is that this new fan is quite a howler. It sounds like a small hair dryer shrieking inside the case. On the other hand, you can easily tell that it won't be frying the electronics :)
I was very impressed with the 2247A, so I added it to the bench collection! The left oscilloscope is the 2247A I just refurbished. Top right is the 2225 I recently calibrated, and bottom right is the awesome 2465A. I love them all...
Dark shot
Edit: Just checked Ebay pricing for 2247 Oscilloscopes. There are a couple of categories. "Pre-owned", and "Parts Only" - meaning that it doesn't work or has issues. The Pre-owned start at $200 and go up to $500. I'd say getting nine Tektronix scopes for $200 (even if they need a little work) was a steal!
No comments:
Post a Comment