Vintage Synth Repair

The old electrical engineers vote, in agreement with Scott from Ohio, to respect the original design intent and replace the tantalum caps with tantalum caps.  In the old days of the 1960's and 70's there were issues with tantalum parts shorting out and blowing up and stuff.  Pretty sure these issues have been worked out OK.  I use many of them for supply decoupling by the thousands in lab equipment I have designed, as well as when re-capping old oscilloscopes/test equipment, and would do so as well in ones I run into inside electronic musical instruments.  All this said, I can't say that I have done a detailed comparison of ESR vs. frequency between low-ESR aluminum parts and the tantalum caps.

Couple of other comments to other posts: Multi-section can capacitors are being made brand new, and are generally available, as previously noted, at tubesandmore.com

Fine to go up in capacitance value, and voltage.  However, with voltage it isn't generally a good idea to go way up in voltage.  Like to use a 450V section when the original is a 50V part.  If you can't find the exact match, put in a discrete part in the event your can has a 50V (or whatever) section.  Extra sections in the can, assuming the voltage is OK, can be paralleled with other ones.

I have found, I think when re-capping a Juno 106, that they used a lot of 16V electrolytic caps on the 15V rails.  I'm not comfortable at all with this and recommend using 25V or 35V parts instead.

Note that newer aluminum caps are usually much smaller than their vintage counterparts.  Therefore, if you want to get some economy of quantity buy, you can, for example replace all the 10uf and 22uf parts with 22uf parts (for supply decoupling, anyway).

For the best reliability, try and get 105 deg C caps instead of 85 deg parts.  Across the board (no pun intended).  Not that you should lose sleep if you can't find a particular value/voltage in 105 deg, but good to try.

Mouser and digikey are great sources for parts.  Online part searching, and ordering all work pretty well (although the printed catalog is good for browsing).

Great discussions!

Hugh




--- In vintagesynthrepair@yahoogroups.com, Daniel Forró <dan.for@...> wrote:
>
> Thanks for your perfect explanation!
> 
> Daniel Forro
> 
> 
> On 9 Apr, 2013, at 2:03 AM, 65 Lotus wrote:
> 
> >
> >
> > Internally tanatlums look like a sponge where their charging surface  
> > area is exposed in 3-dimensions, whereas electrolytics are a couple  
> > of wound, flat pieces of foil.
> >
> > Due to their inherent structure, tanatalums charge very quickly as  
> > the 3-D surface gets easily flooded with electrons, whereas for  
> > electrolytics, the electrons have to spread around the plate to be  
> > saturated. That takes time.
> >
> > So, tantalums are very good for high frequency cap duties, like on  
> > the crystal (clock) circuits. For power rails, electrolytics are  
> > fine of course, but for anything that's a high frequency circuit,  
> > I'd stick with a tantalum if was spec'ed that way to begin with.
> >
> > Tantalums are very unfogiving of overvoltage which is what usually  
> > kills them. Once they're saturated, the slightest voltage over their  
> > rating will do them in, where they melt into a short. They work  
> > great when used properly in a circuit where the probability that  
> > they'll ever see a spike is nil. Replacing a tant with one with a  
> > larger rating will slow down the saturation time, so again, I'd just  
> > replace it with the original rating component. I like the Vishay  
> > brand which is what we use in the aviation industry where I work.
> >
> > I'm no expert (I'm a mechanical engineer however), but I spent a few  
> > hours researching the ins-and-outs of tants one day right before I  
> > did a full recap on my OB-Xa. If it was tantalum, it stayed that way.
> >
> > Scott in Ohio
> >
>