1. Things you need to know about the new ‘Conversations’ PM system:

    a) DO NOT REPLY TO THE NOTIFICATION EMAIL! I get them, not the intended recipient. I get a lot of them and I do not want them! It is just a notification, log into the site and reply from there.

    b) To delete old conversations use the ‘Leave conversation’ option. This is just delete by another name.
    Dismiss Notice

Mil Spec Ag-Pd Tantalum Caps anyone?

Discussion in 'd.i.y.' started by Jonathan, Jun 10, 2019.

  1. Jonathan

    Jonathan pfm Member

    I know MisterC6 was fiddling with mil spec tants a bit - and I've been looking for some decent caps to put into a soon-to-be heavily modded NAC62 (will be doing it in stages) - but has anyone tried these or put them through their sniff detector?


    also - is there any particular trick to accessing the bottom of the NAC62 circuit board? I thought of desoldering one side then 'flipping out' the other but I'm not even sure about that since I'd never tried it before. Just hoping to profit from the experienced.
  2. colasblue

    colasblue pfm Member

    They're only 10uF so you can't use them as feedback caps which is where most people would want their "bespoke" colouration adding cap.

    To get to the bottom of the board you just unbolt the heatsink from the sled. You will need some new heatsink paste when you put it back.
  3. Jonathan

    Jonathan pfm Member

    heatsink? we're talking NACs here - preamps - i do not believe there is a heatsink per se? just four screws attaching the circuit board (but lots of stiff wires to contend with)

    but no - i would be using a 22µF cap in the feedback position with a 20-22K resistor ... to compensate somewhat for removal of the balance control. the value here is arbitrary. Those caps are available in different values.
  4. Jonathan

    Jonathan pfm Member

    i thought i might try either THAT or a Wima MKP4 for the feedback cap
  5. colasblue

    colasblue pfm Member


    Brain out of gear, most people use mil spec tants for feedback in the NAP circuit if they use them at all.

    for the 62 you take the knobs off and undo the nuts that hold the pot and selector switch to the front of the chassis, and take out all the standoff bolts. There will then be enough wiggle room to slide the board back and lift it up without undoing all the connections at the back. If it's CB you also have to undo the nut that holds the tape monitor switch in.

    Not much point in putting tants in to coupling positions to add colouration and then using a film to eliminate it in feedback I'd have thought.
  6. Jonathan

    Jonathan pfm Member

    ah ok - that's the answer i was hoping for! thanks (re: removing the board)

    well it's not clear what i'm putting where - i'm a bit confused at this point - but after reading hundreds of posts about how great the oscons were - but then NOT (after) and Silmics etc ... it wasnt' clear what a good path might be but film or something more exotic like a wet tant in the feedback position seemed like they might be worth a try. as for the coupling caps and tant replacement i'm just not sure
  7. colasblue

    colasblue pfm Member

    The correct os-cons for decoupling are now unobtainable so no longer an option. They were never liked by many for feedback

    I rather like Silmic II's for decoupling but they have a very long run in period (couple of months at least) so no "instant hit" there. I use 47uF Rifa MMK's for feedback but previously also used the silmics

    Current fashion is the Nichicon Muse KZ for both

    all of the above relates to the 35V electrolytics rather than anything else. (axial for decoupling made by Sprague, RoE or SLCE depending on vintage and RoE maroon for feedback. Any plug in cards use RoE maroons)

    Not heard of many people playing with the 10uF coupling tants other than to swap them for films.
  8. Jonathan

    Jonathan pfm Member

    well again ... the 10µF value had nothing to do with the application - but rather for another one. let's just pretend you didn't know the value ... :) thanks
  9. Dougunn

    Dougunn pfm Member

    'Mil-spec' means nothing other than a specification set by the military. It could be a low (cheap) spec for some mundane application or an unusual spec for something extrmely specialised. Either one is no guarantee it will be good for audio.

    Its a phrase beloved of marketing; without clarification it means nothing.

  10. Chris

    Chris pfm Member

    Out of my depth here cos I know nothing about electronics or Naim but I have just fitted some Russian K73-16 caps in my phono pre substituting some US teflons that I was well chuffed with and they are really special. They are Mylar or polyester or something but don´t hold that against them: Transparent across the whole range, much greater detail and informative. Can´t fault them. They are also pretty small so size shouldn´t be a problem. The highest capacitance I have seen is 22uF but voltages from 63v to 600v.
    mik_rik likes this.
  11. Jonathan

    Jonathan pfm Member

    yes but like all names for collections of objects with similar characteristics - it helps you narrow down what you're talking about ... no? let's say we were talking about ceramic disc capacitors for example ... doesn't that give you a better idea of what one might mean?
  12. Dougunn

    Dougunn pfm Member

    Not sure I see your point. . . a collection of objects that conform to an unspecified military specification of some sort . . how is that useful?

    What would actually be useful to say would be what the specification actually is e.g within a defined performance range, guaranteed operating life, etc etc.

    The assumption that because a spec is 'military' it must be good for all applications is fallacious. My father was a research physicist who did a lot of work developing the technology that enables undersea fibre optic cables - the 'spec' for components used in cable amplifiers (or repeaters as they are known) was incredibly demanding. Rejected components were sold on for use in aerospace and satellite applications!
  13. sq225917

    sq225917 situation engineer

    I have some 47uf 50v wet slug tants purchased exactly for modding naim cards, surplus to requirements. £10 each.
  14. Arkless Electronics

    Arkless Electronics Trade: Amp design and repairs.

    Much of the cost in mil spec parts can be due to the degree of testing that's carried out and not due to any special build quality. Not always of course and some parts for specific purposes are indeed about the best you can get.

    I've known parts which are bog standard and also of an old spec that's not as good as current standard consumer spec (due to improvements in manufacturing technology) still be the only one's which could be fitted to certain military equipment simply because there is an in date mil spec certificate for the old parts and none for a better civilian replacement part.... MELF resistors springs to mind from one defence electronics job I was in...
  15. rontoolsie

    rontoolsie pfm Member

    In my CDS3, several of the $1.00 10uF tant coupling caps were replaced with some $80 20uF wet tantalum ones (the 10uF had measureably worse LF performance and phase preservation than 20uF) and I would say the non-trivial added expense was worth every dollar. The AVX caps that Naim use have the benefits that they are pretty close to spec and have long term stability and do not sound that bad (and indeed even contribute to their signature sound), but with the large number of them that are liberally sprinked through pretty much every one of their electronics, it would cost prohibitive to replace them with the wet counterparts. Swapping between different caps is the equivalent of 'tube rolling' and the preferred type is often due to user preferences. But occasionally there are variants that are better in every way.
  16. Arkless Electronics

    Arkless Electronics Trade: Amp design and repairs.

    Personally i strongly disagree that choice of capacitors is even important 90% of the time. Certain capacitors in certain positions and yes there are very worthwhile improvements from having the right ones but if anyone wishes to try simply placing all sorts of caps (of the right value!) in line with say the signal from the output of their pre amps then I think most people would be shocked to hear how little, if any, difference there often is. Yes I've carried out this simple experiment, years ago.
  17. russel

    russel ./_dazed_and_confused

    As I have a few bagfuls of wet slug tants, I respectfully ask you to STFU :)
    Arkless Electronics likes this.
  18. sq225917

    sq225917 situation engineer

    Most of the caps don't matter, feedback caps in Naim circuits make a big difference due to the crap psrr of the circuits.
  19. Jonathan

    Jonathan pfm Member

    the point was that in referring to a 'mil spec tantalum capacitor' that one has a pretty good idea of the kind of component that one is referring to - as distinct from some other type. I was only referring to those as Malcolm seems to feel he has had good luck with those. As i mentioned previously - whether it's a 'mil spec tantalum' or whether it's a 'PET film cap' or whether it's an 'x cap' or even an 'elna silmic' is not material in terms of a value judgment but simply a convenient way to refer to a class of object. We could well be talking about 'stainless steel egg beaters' ... things need names in order to distinguish them from OTHER things ... otherwise dialogue would be useless I would think ... I suspect you're getting a bit wound up assuming there's a kind of implicit value judgement - which there isn't. Though I suppose i might assume the 'mil spec' stuff to be more reliable than other 'non mil spec' items. that's why specifications exist no?
  20. Jonathan

    Jonathan pfm Member

    yes it was the feedback caps i was mostly concerned with ...

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice