1uf Elec Vs 1uf Metal film

[yeeshkul]

Ronsonic can you PLEASE provide us with some samples of the circuit with electro / film caps?

[Dai H.]

don't know if this would be true for all situations, but it seems you may need to compensate and tweak the value when swapping alu electro for film--or at least it has seemed that way in my limited experience. For example swapping the two 10uF BP alu electros for polypropylene films at the input in a Tascam 488mkII seemed to give too much low end and cause some sort of distortion until I backed off the value to 3.9uF(I suspect a little bit more uF would be better in terms of matching the original freq. response but cheaped out at the time). Another way of thinking about this is that in order to do an accurate comparison, the film and alu electro values may need to be different due to better/worse behavior as caps at different frequencies, ESR specs or whatever it is causing any differences.

[R.G.]

Tolerance - the tolerance on films is not worse than +/-10%, often 5%. Aluminum electros are often +80/-20%.

[Gus]

When I "think" I hear a change I often will measure the cap for value etc.
As others have posted tolerance 

Also think about this, the RAT with the 47 ohm resistor The  ESR  of the 2.2uf cap could be what some hear when using a Tant for an Al electro.  How much will the RC change with the high pass step?

[mojo_hand]

I always use film caps in the signal path, for reasons quite unrelated to mojo.  Whatever one thinks of the *sound* of tantalums, ceramics and electrolytics, from an engineer's perspective they're badly flawed.  A filmcap that's slightly damaged will repair itself.  All an electrolytic will do is to slowly dry out.  Filmcaps are typically much closer tolerances, so if you stick a 1 uF into both your input channels, you don't have to worry that one will roll off everything below 40 Hz, while the other one kicks in at 120 Hz.  Filmcaps don't go way off value as temperature changes, and are impervious to moisture.  They tend to be far more linear, whether or not you happen to think that's a good thing for the sound.

From a mojo-agnostic perspective, the only normal reasons to *not* use filmcaps is that they're bigger and more expensive.  Tantalums leak current in a predictable way, that might be useful in biasing a particular, very unusual, circuit.  (I once worked on a Nelson Pass designed class A hifi amp that used that feature.  AFAIK, that was a unique example.)  And there is probably some circuit which can exploit a ceramic which is +22/-82% of value depending on temperature.  But I've never designed a circuit which used any of the technical flaws of capacitors, and, with the exception of subjective sound mojo issues, I doubt that anyone reading this has either.

Most of the above also applies to metal film resistors versus carbon types.  The only reason they are used in any consumer product is because they're dirt cheap.  The only reason they're used in any amplifier is because they're less linear in response to voltage changes, i.e. their value changes as the voltage across them changes.  This makes them a handy, and incredibly cheap, way to add some compression, IF you can find a batch with manufacturing standards that are nearly as bad as they were 50 years ago, which is doubtful.  If you use them in input stages, all you're going to get is a much higher noise level, worse tolerances, and resistance changes with temperature, humidity and age.

So I build my stuff as if I were making it for NASA, and don't worry about the mojo.  I like knowing that, 50 years from now, if what I made has had its power supply recapped a few times, everything will probably be in perfect working order.  If someone buys something I made, and finds the sound of filmcaps and metal films offensively linear, they are always free to replace them.

[Dai H.]

good point (re: tolerance and measurement), I've got some old caps that are waaaay off tolerance--some ridiculously so (an old British 500pF cap that measured something like .02uF) and while ago I experienced a "tone change" in a Fuzz Face (eventually traced to an old .01uF actually measuring .02uF) come to mind. I went over to see if I could find the two 10uF electros I removed and (I think) I found them (two "Tracon" 10uF 100V polarized not BP--I'll have to look at the other channel to be 100% sure these are the ones) and measured on my cheapo cap meter and they did measure 10uF(10.22, 10.33uF).

re: the idea that tantalum and alu electro sounding too poor to use I personally don't buy this. I'm sure there are already many many recordings of hits people liked and thought/think sounded good that had their signal go through tantalums and alu electros. Example: A couple of weeks ago I was searching for info on the Helios console since I was putting together a GGG octavia board(for those that don't know, the driver section of the octavia/o before the transformer appears to be a direct copy from the Helios), and on pictures of  original boards some used a tantalum for the 0.1uF. Another one: SRV and Eric Johnson (two famous gtrists. w/many fans trying to emulate their sound) used original TScreamers. Of course these had .22uF tantalums. So were their tones (when they used the TS) ruined because of the tantalums? Also looking at pics of expensive consoles, it does appear that alu electros are used. It's probably like lots of other things where it depends how they are used more than if.

[chris1001]

I have noticed that on many of my pedal builds/mods, that Tantalum caps often make the pedal slightly more 'noisy' than Panasonic stacked metal film caps, the difference was quite noticable - although not huge, even to my wifes untrained ears!

[mac]

I have noticed that on many of my pedal builds/mods, that Tantalum caps often make the pedal slightly more 'noisy' than Panasonic stacked metal film caps, the difference was quite noticable - although not huge, even to my wifes untrained ears!

My mother can still hear the difference between a good electro and a poly/ceramic/tant cap. She does not like electros. Considering that she is 74, has age earing loss, loves classical music, and I'm a latin son... suficiente para mi!!   

I do not like electros because they sound dark compared to a film/ceramic/tant, and sooner or later need to be replaced. I'm considering to build a pcb to hold 20 x 4.7uf polys to replace the 100uf electro x 500v at my amp PS.
Nevertheless I found that for my rangemaster a cheap electro bypassing the emiter resistor was the best choice to get that vintage sound.

mac

[Gus]

Mac
My laney VC30 has SCR polypros in the power supply

http://www.angela.com/   is were I bought them

I used 1/2 the Al cap value.  You can get then from .15uf to 39uf at 630VDC at that site

[mac]

My laney VC30 has SCR polypros in the power supply

I have a LC30. The original Samwha electro caps are decent, only at very high gain settings it gets a little noisy. IIRC the pcb allows to add the caps in parallel and in a separate board, no need to remove the electro at all.

mac

[jakenold]

It's a matter of taste!

Basil, salt and oregano are great spices, but will ruin your food if you add too much of one of the ingredients. I like electros in some vintage circuits, as I like ceramics, as they add some "dirty noise" that I like and find makes it a little cozier sounding.

It's a feel-thing. Experiment for your own sake, and get some miles of that breadboard!

Jake

[Ronsonic]

Ronsonic can you PLEASE provide us with some samples of the circuit with electro / film caps?

Sorry, been away awhile. Life and cyclocross (don't ask, it's a sickness) season intervened.

I think you're looking for some other -sonic. Probably Soulsonic or maybe even Red House. They both seem to have applied a pretty high level of comparison testing.

Anyway, a lot of very good points throughout this thread. Humans have been using electrons for music for about a hundred years now and there's a lot of ways to go about it. It's pretty easy to test this stuff - or ignore it and focus on other elements of mojo. So grab clip leads or master the ancient and honorable technique of soldering parts in and out of a circuit with a guitar on your lap and the amp hot and see what works for you and your builds. There are so many variables that we really can't tell one another what is best, but we can easily enough find what is best for ourselves and the things we build.

I do want to comment on the question of whether "it makes a difference while rocking a bar with skitchey wiring and trying to get The Cute Chick (there's always one) to dance (FTW).

If you've ever sat in the back of the studio trying (pretending) to read a magazine* while you hear your whiney, unhappy guitar lick over and over again as they dub in other tracks you'll know that careful, living-room level listening in minute detail is actually a very valid test.** It does happen that you need gear that can withstand close scrutiny. Don't you want the A&R guy or whomever to want to listen closely and be drawn into what perfection we may achieve. Even when that perfection is a buzzy, bellowing, wheezing artifact of Chuthulhuan geometry and starved circuit sickness.

Ron

* Yngvie Malmsteen interviews are always the best. The unintentional Spinal Tappian glory of a man whose every note can be predicted by the key signature ripping on other players is priceless.

** When that happened to me I mentioned something and the engineer changed something the next time around without saying anything. Ahhhhhhhhh. The difference was night and day. Good engineers are a wonder. Left to my own devices I probably would've clobbered it instead he just took off 1.5dB at 2K and added a very faint reverb and it all just relaxed and fell into the mix. 

[brett]

So good to see this thread back.  Not.

Maybe I can cover all future posts with this approach:

I (you) know for a fact (because it's what I think  ) that (insert your least favourite cap here) sounds floaty/bloated/wimpy/emasculated/dark/harsh/grainy/weak/...

To the lounge with this rubbish please!

[Dai H.]

I was playing around with my new LCR meter and something I noticed is that a bunch of 1uF alu electros measured about spot on when the lowest freq. (100Hz) was used, while at 1kHz the cap. was a bit lower. Maybe that says something about part of the perceived difference?