What are the best MUSICAL sounding Capacitors to use or recap with in Pro Audio Gear?

[lassoharp]

Originally Posted by JohnRoberts

I recall one esoteric tube amp a few decades ago that used a ton of output tubes in parallel to eliminate the need for an output transformer. Again, not sure why, make it sound less like a tube amp?

JR

Bruce Rozenblit built one and published results in Glass Audio. In his words he did it "Mainly because I was curious as to how it would sound". Probably the challenge from a design perspective intrigued him too.

[cathode]

Originally Posted by JohnRoberts

Deane was responsible or at least very influential in the classic transformer designs that were sold under his name.
JR

Deane was my friend and I will go to lengths to protect his legacy. However, he sold Ed Reichenbach's transformers, the designs were mostly Ed's. Deane seemed more focused about op-amp design than coils, he loved to quote Tobey. Deane is to be credited for successfully marketing nickel core line level output transformers. He one handed me a 1:1 to try as a phase splitter; told me to handle with the utmost of care as the laminations came from some rare batch of nickel that mechanical shock would harm.

Not only did he do the famed 990, but also the 918 which is much more compatible in 2520 applications compared to the 990. The 990 was more of an exercise to exploit the low noise capabilities of the LM394 rather a universal audio gain block.

[TRJanuary]

I have an Orange Drop that can play harmonica.

OK sorry.

[JohnRoberts]

Originally Posted by cathode

Deane was my friend and I will go to lengths to protect his legacy. However, he sold Ed Reichenbach's transformers, the designs were mostly Ed's. Deane seemed more focused about op-amp design than coils, he loved to quote Tobey. Deane is to be credited for successfully marketing nickel core line level output transformers. He one handed me a 1:1 to try as a phase splitter; told me to handle with the utmost of care as the laminations came from some rare batch of nickel that mechanical shock would harm.

Not only did he do the famed 990, but also the 918 which is much more compatible in 2520 applications compared to the 990. The 990 was more of an exercise to exploit the low noise capabilities of the LM394 rather a universal audio gain block.

Thank you for the extra background information. I recall a personal conversation with Deane back in the early '80s where he talked about having a sister who was a chemistry (?) major, and that was where he got his insights about transformer core metallurgy.

I probably ASSumed he was more involved in the transformer design than he was. I was talking circuit design with him and transformers just came up in passing (wrt how to properly drive output transformers). Even back then I was already pretty much transformer-less input and output.

Yes the 990, is a classic design and the inductors in between the input long-tail-pair is brilliant. RIP

JR

[Piedpiper]

How well does the VMP-2 compare to other modern hi end tube pres like the Manley, etc.?

[Thee_Adleen]

caps are caps are caps

it's not what you use
but how you use it

if the cap doesn't work
find a different components to use
even if everyone says "that's ********"
don't reinvent the wheel
make something better
invent the sphere

don't be limited to linear movement
and linear thinking
period

[foldback]

I spoke with Deane Jensen many times on the phone, I still have my 70's Jensen Application binder too :-)

In the early 80's I was building mic preamps, I was already on a quest for Neve quality I could build myself.

Deane did not like Neve, every time it was mentioned he would diss the sound as distorted. He sent me to his output transformers (which were really Ed's) and they were made with nickel instead of steel laminations. Deane and Ed both had a lot of experience with steel lamination material at API (which I tend to think of as the East Coast sound). The nickel core transformers have significantly less group delay and phase smear.

Deane also did not like the 115 mic pre transformer, he thought it was far too distorted sounding due to the 1:10 step up ratio. He much preferred a one-to-two step up with the balance of gain coming from electronics. He loved the isolation a transformer provided but he was hell bent to get rid of distortion.

After I built a few mic preamps I tried a Hardy and decided I'd rather spend my time writing and recording than being an equipment manufacturer.

Good music to all!

[fnninns]

When you talk about "west coast" or "L.A." mic pre designs you really need to look beyond Mr. Jensen and look at the designs that John Hall developed.
His designs are used in Electrodynes, Quad8s, Spheres, etc...

Although Mr. Jensen took things a few steps further and applied a lot of thought he was still building on previous designs.

As far as caps go... the Panasonic FMs , FCs, Vishay (essentially old Spraque) and the better Nichicons are all adequate in almost all audio applications.

These are all 105C caps and once you convert centigrade to fahrenheit and look at the operating temperature environment of the equipment the caps will go into you see that the 105C caps will last A LONG LONG time.

When you talk about removing coupling caps you also have to remember that manufacturers didn't inculed them out of ignorance.
They are there because in certain situations they ARE neccesary.
It is important to look at a console as a large system and not only look at one circuit at a time.
In a stand-alone situation you might be able to eliminate coupling caps, but in a complex system like a large format console being used in all types of environments the coupling caps probably do serve a purpose.
Also, if they were not neccesary most manufacturers would not have included them because of cost alone.
I am very familiar with MCI console designs and I can promise you that if MCI could have eliminated the coupling caps in a JH-500 series console they would have.
As an example:
A JH-538C has close to 6000 electrolytics and at least 25% are coupling caps.
If MCI could have eleiminated 1500 caps they would have.
In a JH-500 series the coupling caps are mostly 100 uf and 200 uf and they are some of the more expensive caps in the console.
Eliminating them would have saved more than 25% of the manufacturing cost relating to caps.
If you follow the eveolution of MCI equipment and talk to ex-MCI designers you learn that keeping the product affordable to people in music production was their main goal.
The consoles were not over-built like some audio gear (ex: Ampex) because MCI did not try to go after military/governmental contracts.
MCI built a product that was as functional as possible and still affordable and in the quest to keep prices and profits down they simplified many designs.
They kept most of the coupling caps in all of their designs because they are needed in most cases.
This same principal appiles to most audio manufacturing companies that don't build products for the military although there is obviously a boutique market for "over-designed" or "premium grade" products.

Personally, I don't like to second guess the people who designed equipment.

[spice house]

I'd bet if you were more knowledgeable about electronics than the designers, and you weren't limited by penny pinchers, you couldn't help but 2nd guess the designers. You'd of course forgive them, though, because they are trying to make a profitable, functional, and marketable product, not the most amazing sounding thing ever.

[Piedpiper]

no matter how expensive a product, there is always a budget to adhere to... and there are also always improvements to be made by throwing money and smarts at it... if one wants to prioritize quality...

[jrp]

So what about modern analog consoles? Or other analog gear?
I get the impression that more and more manufactors promote their gear as dc coupled.
So the use of modern precission opamps is no secret, if they had been available 30 years ago someone might have saved 1500 caps and put the money in better amps...(?)

On the other hand, i tried 2132 P Opamps in my console, they are supposed to bring 0.5mv offset. Didn´t measure, but the switches do produce more clicking noise than with caps.

But, when i picture my conole, there are many many double caps, like one before, one after the switch, fader, etc.
Why would someone do that?

[javd007]

I'm putting the thread on Caps Lock...

[jrp]

So offset can not only come from the output, but also the input of an opamp stage?
I always thought of getting dc from a pot by placing the cap only in front of the pot when it is followed by just an opamp.

[Henke]

Read up on input bias current. Fet input opamps are generally better in this regard.

[jrp]

i try... but it´s hard if you have no technical background. Until now I sort of build circuits like lego, always assuming things were ideal.
The Opamp Output having a dc offset that needs to be blocked makes sense, but what goes on on the input seems a different beast...
Reading on wikipedia i figure that output errors are caused by input currents, but that would not explain why a cap is needed on both sides of a pot.

[JohnRoberts]

Originally Posted by jrp

i try... but it´s hard if you have no technical background. Until now I sort of build circuits like lego, always assuming things were ideal.
The Opamp Output having a dc offset that needs to be blocked makes sense, but what goes on on the input seems a different beast...
Reading on wikipedia i figure that output errors are caused by input currents, but that would not explain why a cap is needed on both sides of a pot.

An opamp with input bias current directly connected to the wiper of a pot that has a changing DC resistance to ground (when rotated) will result in a changing DC voltage as predicted by ohms law (V=IxR).

Many modern opamps have very small bias current from use of FET input devices, or even first order cancellation circuits inside modern bipolar opamps.

JR

[jrp]

Ok. I can see that i was onto some mistake once more. For me, naivly thinking, the bias current was the current that flows into the input stage, determined by signal voltage, source impedance and if present a resistor parallel to the input.
If i do get this right, i rather have to imagine current flowing out of the input, into the pot, leaving a voltage depending on the pot setting.
Sorry for using this thread for my ignorance, this should be about caps and not teaching me basics...

[JohnRoberts]

Originally Posted by jrp

Ok. I can see that i was onto some mistake once more. For me, naivly thinking, the bias current was the current that flows into the input stage, determined by signal voltage, source impedance and if present a resistor parallel to the input.
If i do get this right, i rather have to imagine current flowing out of the input, into the pot, leaving a voltage depending on the pot setting.
Sorry for using this thread for my ignorance, this should be about caps and not teaching me basics...

Nah it's all the same poop...

Input bias current is a constant amount and flows in or out depending on whether input devices are PNP (out) or NPN (in).

JR

[jrp]

of course!
Ground is in the middle of the rails, it doesn´t matter if current goes in or out.

Well it does, but not if we are talking about switching or pot noise!
Naive thinking once more.

[mandoblast]

Anybody who says caps don't matter should have the experience (maybe just once) of hearing audio coming from nothing more than the electrolytic coupling caps on a board... "Look, Mah! Electrostatic speakers!!" I suppose some electro-mechanical loss is inevitable with any capacitor, but seriously, is that were you want your signal going?

[TC Custom Audio]

Get some Panasonic FM and be done with it. Or you can waste a bunch of your time looking for more exotic stuff.

[JRguitarguy]

Dude for power supplies the ELNA and black gates are amazing I have a stash of them xD

[JRguitarguy]

and I got bunch of Panasonic caps from junk over the years lol

[spice house]

Originally Posted by Frank_Case

Electrolytics:

The best current substitute are the Panasonic FR series caps available from Mouser and Digikey. Just great sounding caps with twice the expected life of the older FM series. I noticed that the FR series is about 35% higher in price these days then the FM series which is still being made. Don't scrip for pennies and instead pay the difference because FRs are worth every cent and they still are down right inexpensive for the quality you get. I find that bypassing large cap values (220uf and higher) with a 1uf polypropylene cap helps. Bypass smaller FR values with a .1uf polypropylene for best results.

Sweet, didn't know there was a newer series, going to check those out

edit: here you can compare specs
http://www.digikey.com/product-detai...411-ND/2433545
http://www.digikey.com/product-detai...2923-ND/762833
http://www.digikey.com/product-detai...1547-ND/589288

... the things I do while bouncing down...