ESI DEKABOX DB655 Repair Help

[MikeLogix]

Greetings; I saw that there was a DB655 tear down thread already posted and contemplated asking my question on that post but I worried that I would be thread hijacking. I therefor decided to create a new thread.

I was up late one night and made a poor Ebay purchasing decision. I was wanting a KVD type Dekabox, but what I purchased is more of just a pot with no center wiper On top of that, the unit I bought was received broken. The seller said "did some basic test and appears to be working"
Well, as it turns out the above statement was false, as I received the unit and the inner dial felt like it was jamming up, I gave it a few turns, and then heard a kerplunk, and then there was something rolling about.
I went back to the Ebay page4 to see what my options were and I saw at the bottem, No Returns, sold as is/need repair. So I am stuck with the unit and lesson learned.

Anyways... The problem is on the left half (see image below), inner dial. I opened the left unit and a ball bearing was rolling around inside.

Thus far I can see that I have three problems with the top section which is the 100k part.
problems
 1. Ball Bearing needs to be reinstalled
 2. A small plastic part (part of indent I think) is loose, looks like it will fall off.
 3. A lock type nut came un-threaded.
My question is, I want to repair the unit but not sure how to proceed as I don't want to cause further damage due to ignorance so I thought it best to see if I could get advice on the disassembly process. I do see the two nuts at the top but was not sure if the whole thing would come apart upon removing the nuts. So any advice on repairing this thing would be great. My other concern is that the resistors appear very fragile, so I am hoping I don't have to do any soldering.
Below are images showing the problems.

 

[Vgkid]

Ill take a look at mine either tonight/ or on Saturday. At least I will post pics tonight.

[Macbeth]

I was up late one night and made a poor Ebay purchasing decision. I was wanting a KVD type Dekabox, but what I purchased is more of just a pot with no center wiper On top of that, the unit I bought was received broken. The seller said "did some basic test and appears to be working"

For future reference the esi item you should have been looking for is the DEKAVIDER RV... series.

That's not to say you have a poor DEKABOX resistor (when you fix its mechanical faults) but really the lack of an input and output should have been enough of a clue? 

I hope it was cheap enough and you restore it anyway! 

[Vgkid]

The compact voltage dividers are of the DV(411/412) type. The easiest way to remove the resistors  is to use a pair of hemastats(they have other names/needle nose pliers) and heat the solder up until it melts. Then pull up the lead out of the solder blob. You can do a whole decade in under 10 minutes.

[MikeLogix]

Well, I been meaning to post a follow up here for awhile, but busy with things and all. Anyways, I did decide to tear into my Dekabox, and I managed to repair the unit so I thought I would share the process.

Tools Used:
Flat blade screwdriver
1/4" Socket driver
1/16" Ball Allen (Used for dials, see image below)
Crescent Wrench
Solder Iron (set to 380C)
flux
--------------------------------------------------------------
Being old, I found myself challenging my aging dexterity with this project, I had to finagle myself in strange positions to get my solder iron at just the right angle as to not inadvertently burn or melt the delicate fragile looking resistors set around the wafer switch assemblies.

First thing I did was to set all the dials to zero just to make things easier at the end. Using the Ball Allen I removed all the dials, then desoldered the two wires on the chassis banana jacks.



With the dials removed I used a wrench to remove the nuts securing the two resistor switch units to the chassis, and pulled the two resistor assemblies free of the mounting stand or whatever else you want to call it. I removed the housing of the broken resistor and had to un-solder one wire connecting the assemblies together:



Next step was to loosen the 1/4" nuts securing the whole switch assembly together. I discovered that the nuts were locktighted to the screw shafts, and as I turned the nut, the whole threaded shaft would work its way out. To overcome this, I threaded on a cap nut which allowed me to hold the threaded shaft in place while loosening the locking nut.
At this point I also had to desolder two wires attached the top wafer switch.


With the nuts and wires removed, it was easy enough to pull the top wafer free.


I then slipped out the damaged shaft assembly, the metal ball bearing and plastic indent had fallen out.


While I had the shaft out, I noticed the shaft was bent. This was not expected and I was not sure how to get it true again.




My vice just happened to have a groove in it that was just the right size, so using the vice I was able to get the shaft almost back to straight, not quite but almost and I hoped enough.


I then reinstalled the plastic indent piece and ball bearing.


And then I had to press fit the locking spring washer (or whatever it's called) on the other side. I used a deep well socket for this part and used my drill press to mash it in place.


Once this was done, I was ready to reassemble everything and test.




The reassemble was just a matter of reversing my process above. With everything back together, I measured the resistance with all the dials at 0. I got about 30 milliohms which is out of spec, so I will have to revisit that problem at a later time. I zeroed out the error just to check the resistors, they are still pretty spot on. 1 gave me ~.998

All the resistors check pretty spot on, and the dials rotate nicely now. One problem I am still having is that even fully tightened, the two outer(i meant to say inner) knobs slip, not sure the best way to fix it.
That's about it, and maybe this post will help someone else who needs to perform a repair.

[MikeLogix]

I have to apologize. In my last post I had used Google Photos as my file share site, and I guess no one else could see my images, so I re-hosted them on imgur as was suggested by a forum user. So sorry about the mix up, hopefully the images are visible now.

[Vgkid]

Nice job with the repair, good write up as well.
With that bent shaft, that must have happened when something must have forcefully hit the dials from above.
With the loose dial, make sure both set screws are tight. That nut holding the killohm range decade to the faceplate loosened up. So that unit wiggles. I will just leave it, I don't want to realign those numbers.
Your zero resistance seems fine. I have seen 26, and 28 milli-ohms being reported. So yours is in that range.