Recapping a Dimension

Started by gtnicol, December 01, 2013, 02:40:25 PM

Previous topic - Next topic


I have a couple of Dimension boards that are flaky. Given the recent success with MO drives, I guess the logical thing to do would be to recap the boards to see if it fixes them.

Has anyone already done this?

I'll take some scans of boards I have, but from looking at them there appear to be a couple of variations in the capacitors. It also looks like they're not all through-hole...


Many of the caps on the Dimension are surface mount.... I'm thinking of changing them to regular electrolytic capacitors. Any thoughts?


There are many folks that do just that. Just bend the leads of a radial electrolytic and solder them to the cleaned pads. Others swear by chip-style tantalums that don't contain anything that can ever leak. Personally, I prefer the look of the OE electrolytic aluminum cans so I replace them with modern equivalents. The best of both worlds (no leaks with original looks) are modern organic polymers. They have the look of the aluminum cans but no corrosive electrolyte to leak 20-25 years down the line.


I might look into the polymers or tantalums. So long as the values (voltage, capacitance) are the same, it really shouldn't matter which we use...


While we're on the topic, this article may be helpful for some general background on what type of caps to use:



So on one Dimension, we have these.

... and on most of my others, they have these.

On the first one, there are also some 16v 47uf (4 out of 13). I'm thinking of replacing

13x 16v 47uf

with these polymer ones.

Do you think it's worth the extra $$, because the 1x 25v 220uf will probably end up being electrolytic anyways.


This topic reminds me that I still have not finished  my slab re-capping document...  I can send you the current draft.

From your photos I see that NeXT were as generous with the solder pad size on the Dimension as they were on the slab main boards.  This makes the replacement much easier and much lower risk without specialized SMD rework equipment.  A good quality temperature controlled soldering station is all that is really needed.

When using polymer or tantalum replacements it generally would be a good idea to reduce the capacitance value.  This is because their internal impedance is much lower than aluminium electrolytics and the same bypass effect can be achieved with a lower capacitance.  In addition, these capacitors present a lower impedance load to the power supply.  Depending on the power supply design this may actually destabilize the power supply and cause high frequency oscillations.  To some degree this is true with modern electrolytics as well since they tend to have lower impedance than their ancestors from 20 years ago.

Another important aspect specific to tantalums and polymers is the voltage rating.  Unlike electrolytics, they have effectively no over-voltage tolerance and when exposed and not current limited they will fail catastrophically (explosively).  They therefore need to be additionally derated beyond their alu equivalents.  There is lots of useful information available from various manufacturers, such as

On the three slab main boards (one was my original, one from Rob and one from you Gavin) I used modern high quality aluminum electrolytics.  I figured that another 20 years of service life should be sufficient for now.  I have also purchased a complete set of polymers that I have in reserve.  If maintaining the original look is not important then tantalums would be the best choice as they effectively don't age and when properly derated will outlast the useful life of the equipment.


I actually have 2 boards I plan on recapping... so maybe order 1 set of modern electrolytics, and one set of tantalums? I don't think maintaining the original look is absolutely necessary, and it'd be interesting to know if it works.  It shouldn't hurt to use the same capacitance if we increase the voltage right?

I'd appreciate a copy of your recapping manual.


I don't think there would be problem with keeping the same capacitance value.  The polymer caps I ordered for the slab were Nichicon PCV1D470MCL1GS and they are available at Mouser as well as Digi-key.  They are 47uF rated at 20V.

I will have another look through the draft some time over the weekend and I will send it your way...


Ok, I finally spent the final bits of time on the slab recapping document.  I loaded it into my DropBox account and it is available at:

Please let me know if you have any questions or comments.