Another capacitor question.

[zooterman]

Please have a look at the picture, can you advice me what to search for.
( the light blue ones )
Thanks

['LIVEWIRE?']

They look like radial Electrolytic Capacitors to me, don't know if they're a make particularly prone to failure, but, AFAIK, they could be replaced with any reutable make of Radial electrolytic of suitable Capacitance & voltage rating.

[TrevorG3VLF]

How did you manage to get many items in focus but not the capacitors?

Please show us what is printed on them.

[Radio Wrangler]

I think I could make out +/-1% on some of them. That rules out electrolytics and rules in precision polycarbonates or polystyrene.

It looks like some sort of precision filtering is going on, and those capacitors are usually reliable and stable in value. So I don't see any point in replacing them. There's a bead tantalum in the background. More likely to give trouble, I spy a chip in a socket!

David

[turretslug]

Agreed re. the blue ones, pretty certain that they're Philips radial polystyrenes, a slightly more couth presentation than the axial flimsy-lead Suflex etc. polystyrenes. No reason to cast undue suspicion on them.

The orange polarised fella (and it has two leant-over mates in front of it) is a Philips solid aluminium type, again very good- though like any electrolytic-type principle cap, not entirely above suspicion. I have heard it said that Philips knew of the problems many other makers had with tantalums and presented this type as a more reliable alternative with similar low-ESR and good HF characteristics- though it smells of being a "chicken-egg" type tale with possibly confused origins. Certainly, I tend to trust Philips as generally knowing what they were up to with many of their passives.

[ukcol]

I think I can make out "630V" on one of them which also supports the idea they are not small electrolytics.

[zooterman]

Hi Guys,
They are marked in Pf varying values, voltages are 160 to 630 volts, in the process of a repair and just wondered what type they were, iI will try and get a better pic, but I'm not David Bailey.
Cheers

[julie_m]

There were electrolytic capacitors in that sort of presentation, but they long predated the type of resistors in your circuit. Also, I can see "1000 ?F" on one of the capacitors. A 1000 µF capacitor that size is just about believable, but not one that old. A 1000 pF precision cap (polycarbonate, ±1%) of that size is far more plausible. (There are also polystyrene precision capacitors; but they are quite distinctive, in a double-ended, transparent, sometimes blue-tinted encapsulation that clearly shows the capacitor's internal construction). Adding weight to the "precision" idea is the presence of some ±2% metal-film resistors as well as some ordinary ±5% carbon film ones.

Obvious question: What is it? You'd expect precision resistors, capacitors and op-amp ICs in something like a magnetic cartridge preamplifier, for instance ..... probably not in a power supply, though (well, some of the resistors might be precision types). And for that matter, what's wrong with it?

Originally Posted by Radio Wrangler

More likely to give trouble, I spy a chip in a socket!

That's not entirely fair. Whenever I have to remove an IC, I always fit a turned-pin socket for its eventual replacement. Sooner that than cut through each pin and desolder them all one by one again, anyway (and kinder to the print).

Those don't look like turned-pin sockets, though, I'll concede.

[Radio Wrangler]

I could be quite bitter I've got a Marconi 2024 sig gen at work that's giving trouble and needs a number 3 sized clout to persuade it to start up. It seems to be the ROM socket, and it's a turned pin one at that. Next time it's apart I'l splay the legs a little.

David

[zooterman]

Hi Guys.

More pics, They are out of a Studer 810 tape deck, can you identify the blue one marked 68ns?


Thanks, Wal.

[Radio Wrangler]

Probably 68 nano farads. Certainly not electrolytic. Probably quite OK.

Why the need to change? What's wrong with it?

David

[zooterman]

Not saying anything is wrong with the 68ns, but wanted to know what to replace them with if there was.

[Radio Wrangler]

Ah, OK. In the unlikely event.... You could just search Farnell or RS for XXXpf for the lower capacitance parts and search for 1% which might get you a C0G or NP0 tempco part.

One old trick was that some ferrite materials like Siemens N28 used in pot cores like the one on your board, have a temperature coefficient almost the reverse of that of polystyrene film capacitors. This was used to make precision things less sensitive to temperature. You may have to find true 1% polystyrene capacitors to keep original performance.

68nf is not going to be polystyrene.

David

[turretslug]

The 68nF in question looks like a Siemens (now Epcos) multi-layer ceramic. Never got to the bottom of why they used "S", rather than "F", it certainly wasn't mhos! (They liked that thick blue epoxy for some reason, it also featured on their MOVs). They are a pretty dependable component, but with provisos- some of the cards I worked on might have dozens in parallel across the supply for logic decoupling, one would go short/low resistance. My solution was to remove it from the frame and tack a bench supply across the power tracks. Apply finger-tips to them in turn, see which one gets warm. I got a hell of a telling off for this, apparently this wasn't the "proper" way to do it...

[Herald1360]

I suppose if the PSU had too much grunt you might have burnt a track out, but an amp or less should be OK on the power rails. Seems like an elegant way to do it to me.

Zooterman- getting a clear picture of a board like that can be easier if you stand back and use telephoto rather than getting up close with macro. Most cameras now have enough resolution for you to be able to blow up the picture and crop it to get just the area you want with a reasonable depth of field.

[turretslug]

Originally Posted by Herald1360

I suppose if the PSU had too much grunt you might have burnt a track out, but an amp or less should be OK on the power rails. Seems like an elegant way to do it to me.

I was a tad peeved, especially as it meant the rapid and effective turnround of a number of cards that had been dismissed as "bxxxxxds"! I'm broad-shouldered, I shrugged it off as an amusing demonstration of how a few older men of a Certain Background really don't like cocky young whipper-snappers showing too much initiative or imagination... The cards in question were '70's-style glass-fibre double-sided with generous tinned tracks, I always kept the current-limit setting low.

Anyway, to put things into perspective, I saw perhaps half a dozen low-resistance caps in all among a few dozen on each of several hundred varied cards in service, I also saw a few failures of yellow epoxy (Philips?) multi-layer ceramics. It gave me a certain distrust of this type of ubiquitous capacitor- I wondered if, as is said to be the case with the old tantalum bead types, their low internal impedance could be their undoing in PSU/ low source impedance situations. Whenever I used them eg. around IC regulators, I made sure there was adequate input fusing. It's entirely possible that multi-layer ceramics are as trustworthy as anything else nowadays, I haven't done concentrated bench-work for some years.

[Maarten]

I have never seen multilayer ceramics in a round plastic housing. I'm pretty sure they're foil capacitors of some sort, especially given the 1% tolerance and the probable use. I'd go with Radio Wranglers suggestion.

[turretslug]

Maarten, the capacitor under discussion in post #11 onwards was the flattish darker blue one marked "68nS" in the second photo of post#10, rather than the lighter blue cylindrical types in the initial posts- no question that these ones are foil types.

Agreed, I've never seen multilayer ceramics in cylindrical radial plastic housings but cylindrical axial ones in either epoxy or thermoset casings were very common at one time, I encountered them in US professional kit, they also featured as logic decouplers in both consumer and professional stuff over here and many catalogue listings. Certainly not in close tolerance, as you say and cylindrical format kind of went against my understanding of multilayer construction

[zooterman]

So are we agreed that the larger light blue are polycarbonates or polystyrene caps,
Wal