Klein Bottle problem :(

[Govmnt_Lacky]

...dipping my soldering iron tip into some solder paste that I stole from Brian

I don't know if anyone has mentioned this but, THIS is NOT a good idea. Solder paste is highly corrosive and will dramatically shorten your irons tip life.

Just my 2 

[EBK]

...dipping my soldering iron tip into some solder paste that I stole from Brian

I don't know if anyone has mentioned this but, THIS is NOT a good idea. Solder paste is highly corrosive and will dramatically shorten your irons tip life.

Just my 2 

My current tip has been going strong for over 4 years (since my soldering station was brand new).  The previous soldering iron tip I used also did not wear out from corrosion (I probably used it for 10 years or more, and it also came with the iron when it was new).  In fact, I have only once in my life replaced a tip due to corrosion (the crappy Radio Shack iron I learned to solder with).

[benny_profane]

Has anyone tried contacting Peter about the added buffer? He might have some good insight/be interested with changes—especially if he's considering a rev 2 of the board.

[Masochist]

Very nice! Ok I think I get the layout now. Have you had a chance to test it with the KB? Really itching to know if this is viable! Great work regardless man.

[Masochist]

Has anyone tried contacting Peter about the added buffer? He might have some good insight/be interested with changes—especially if he's considering a rev 2 of the board.

I was thinking about emailing him but I might leave that to the actual creator of this potentially excellent alternative.. EBK have you tried getting in touch with Peter?

[EBK]

I haven't even tested this yet. 

I don't mind anyone sharing the idea with Peter, and I don't mind him using it if he wants, but I don't need any credit for it.  It is a trick used in many many other circuits and not an idea I invented.  Peter may have already thought of it but opted for a simpler design for cost reasons.  My fix
requires many more solder joints and creating a new board.  His fix is suitable for his board layout with very minimal modification.

(In other words, I am humble and shy.  )

[benny_profane]

I haven't even tested this yet. 

I don't mind anyone sharing the idea with Peter, and I don't mind him using it if he wants, but I don't need any credit for it.  It is a trick used in many many other circuits and not an idea I invented.  Peter may have already thought of it but opted for a simpler design for cost reasons.  My fix
requires many more solder joints and creating a new board.  His fix is suitable for his board layout with very minimal modification.

(In other words, I am humble and shy.  )

Haha, I'll leave it to you to email him. I encourage you to do so, though. I bet he didn't offer this as a solution because it requires a daughter board, but a board revision could totally implement it. It's a four layer pcb that isn't exactly small...I'm sure there'd be space.

[Masochist]

Nearly finished constructing the board. Is that a 100nf MLCC you list there? I don't think I have one of those bad boys.. I've got some 100nf film caps though, will they do the trick or should I go and source a 100nf MLCC?

Also, have you crossed out R68 and R69 on the diagram for connecting the board because you're saying to remove these resistors first? Thanks man

[EBK]

Nearly finished constructing the board. Is that a 100nf MLCC you list there? I don't think I have one of those bad boys.. I've got some 100nf film caps though, will they do the trick or should I go and source a 100nf MLCC?

Also, have you crossed out R68 and R69 on the diagram for connecting the board because you're saying to remove these resistors first? Thanks man

For something trivial like this, a 100nF film cap is probably ok.  It is just a decoupling cap.  Ceramic caps work better at shunting high frequency noise than most other types of caps.  Try to keep the wires going to the buffer board short though.

If you are going to use the buffer board, then, yes, omit R68 and R69.  The two resistors on the buffer board perform the voltage divider function instead.

[Masochist]

Great, thanks for your help with this!  I'll have it installed tomorrow morning at some stage once I receive my MLCC's and pick up some heat shrink tubing.

It doesn't matter what side I solder the leads to on the resistors, right? I'm just going to utilise the same technique Peter used and snip the existing resistors and solder the leads from the new daughter board to the original resistor leg nubs. Should be fine just to leave the board unsecured do you think? I'll just keep the leads as short as possible.

[EBK]

It doesn't matter what side I solder the leads to on the resistors, right?

I'm confused by this question.  Are you asking about which pads on the Klein Bottle PCB you should attach the buffer board to?

The buffer board wires must go to specific pads on the Klein Bottle board. 

The left side pad for R68 on the PCB is the V+ bus.  This must be connected to the V+ wire of the buffer board.

The right side pad for R68 on the PCB is the Klein Bottle's Vref bus.  This is connected by a trace on the PCB to the left side pad of R69.  The Vref wire of the buffer board must connect to one of these pads.

The right side pad of R69 is connected to the ground pour of the PCB.  The ground wire of the buffer board must connect here (or at least to another spot on the board that is also ground).

I will try to solder mine up tonight and show you what I mean.

If you insulate your board with sone heat shrink tubing like I did, then you can probably let the board float unsecured inside the enclosure.  You should probably secure it somewhere with some double stick foam tape or hot glue to help ensure that the wires wont break from mechanical fatigue eventually.

[EBK]

Actually, I want to test a few things before I solder in my mod board.

1.  I want to measure Vref in the stock version.

2.  I want to recreate the bleedthrough so I can tell if I've improved it after my mod.

[Masochist]

Sorry for the confusion, I obviously still have a lot to learn!! My thinking was that because it doesn't matter what way around you put a resistor, it might not matter which way around I install the wires for the buffer board. Thanks for straightening that out!

Ok, I'll wait until I hear the results from your tests before I install the board. Thanks again!

[EBK]

Test 1.  The voltages I got (with all 3 loops in bypass) were:
9.43V at DC jack
8.82V on V+ bus
4.27V on Vref bus

I was curious whether the DC value of Vref was being pulled significantly lower than half V+.  Peter wrote +12V and +5V on his PCB trace diagram, suggesting he expected Vref to be slightly less than half V+.  This test was also a sanity check to make sure my buffer can handle the load (it can).

A quick calculation shows that the DC load on Vref (the effective resistance to ground in parallel with R69) is about 15.2k, resulting in about 0.28mA DC flowing out of the Vref node from the voltage divider.

With all 3 loops on, the DC current out of the Vref node was about 0.16mA.  V+ was 8.78V, and Vref was 4.31V.  27k load on Vref.


I don't quickly see any place in the circuit where this minor discrepancy seems to matter though, so I'm going to stick with my resistor choices on my buffer board, so my Vref post-mod should be higher.


Masochist, you don't mind me hijacking your thread with my mod stuff, do you?   
I don't want to seem rude, and I can start a new thread if you want.

[Masochist]

I don't mind at all man! I'm highly appreciative of your help and it's been awesome seeing this mod develop! Maybe it deserves its own thread though, as some KB owners might skim over this thread after the first page. What do you think?