Bonehead Cap question

Last one for this week fellas..On this pre I am modding it is reccomended that you double the value (and quality) of the coupling caps if you put a good input transformer in (improves low end response)....I don't have any idea what this means precisely in terms of the rating..I tried googling it before bugging you guys....Okay..today is the trip to Borders to get a good beginners guide to electronic and circuits

Ray

Say, if the original cap is 1yF, you'd use a 2yF (or 2,2yF). Simple as that :grin:

Now 'doubling'/improving the quality is a VERY subjective matter. The pre I'm doing now only have ol' huge Cornell Dubilier oilcan type caps in the signalpath because I'm opting for a specific sound. I would not use them in say a Neve pre. I'd use crappy tantals there. But that wouldn't be good for a Millenia/Avalon type of pre. Nice poplypropylene caps, like Solen or MIT are suitable for that app.

What pre are you modding?

A phat lowend also sometimes comes from the right trannie/output cap combination and upping the values here might make the tone suck..

I am modding a Rolls rp220 that I got for 50$ (so it is really an experiment and practice for my DIY skills since I am a total newbie)..the theory behind doubling the value is changing the low frequency response..with a good input tranny and some changes to the signal path (better opamps, caps, removing the shunt caps) you can change the value and improve the sound..I am assuming "yf" values translate to "uf" values here in the states

Ray

It's a good idea to enlarge the inputcap's value. Use filmcaps throughout the project. I'll be the PSU can do with some beefing up too :wink:

Wouldn't yF be a "yoctofarad?" 1yF = 0.000000000001 pF
The way I figure, a 2.2yF cap could be constructed by placing two straightpins in outer space some 37,300 miles apart. :grin:

I know--a geeky but not terribly useful first post for a long-time lurker, but I couldn't resist, since I just read a reference to the yocto- prefix just yesterday!

I'm in the process of DIY-ing the studio itself. Once that's done, I'll have 30 empty rack units waiting for lots of DIY gear. I've been accumulating parts casually for a LA2A or two and some 1272's.

Great community with a lot of creativity and learning happening. Glad I found you guys and followed you over to the new digs. :guinness: :sam: :thumb:

Regards.

--Bob

Hello & welcome Bob :wink:

yF = microFarad, micro = 1/1.000.000 . A picoFarad is 1/1.000.000 yF not the other way around..

I haven't seen any mention of an input cap..it has a transformer feeding a 12ax7 running at 250 volts..there are some mentions of beefing up the power supply but I can't follow exactly what was done..I could post the mod here and see what you think..thanks.I planned on replacing all the caps I can..

Ray

Sorry 'bout the input cap thing, dunno where that came from :? A schematic would be very handy too. I have a bit of experience with Bellari/Rolls and I'm guesing it's a transformer->tube->dual opamp for balancing, right?

If no schematic (have you tired emailing them) then a photo would be of great help :wink:

Sismofyt -- I don't wanna hijack this thread, but you touched on something thats been on my mind lately.

You mentioned selecting different kind of caps for their tonal qualities. I'm a complete newbie, just accumulating parts for my first DIY (ssl)... I'm using metal film caps except where tant's were specified, cause I heard they are the best in terms of tolerance and longevity -- but might other kinds have given aslightly different sound? Are there any general guidelines when selecting caps for different projects -- I mean do some encourage warmer or bassier or more dynamic tones then others? Any way to tell whether poly, tantalum, or metal film caps are most appropriate? I'm guessing it depends 100% on the circuit they're in (and hopefully that stuff will make more sense to me in time), but figured I'd ask anyway.

Yep, that is the config..no answer from them on the schematic and a web/news search turns up nothing as well..I just got (10 minutes ago) my telefunken 12ax7's to put in..I noticed a large improvement immediately...a little clearer and more present..I am letting them warm up before i do a little singin'..I have a nice digital camera so I can send you as many pics as you want..Thanks so much for the helpl on this..tonight I am going to bypass the caps in the output section rendered useless by the burr brown 2604 op amps..Tommorow I am going to order the replacement caps from digi key and 2 lundahl mic input transformers..as for the power supply here is what I read...

I changed R1 and R28 from 10K to 2K5. This is in series with thetransformer secondary and the grid of the first stage, and thismodification reduces any possible noise from the 10K resistor andmoves the HF rolloff up 2 octaves. I also replaced the shuntcapacitor (C1, C12 = 27pF ceramic) with a 26pF polystyrene to make itsound nicer.Other resistor changes: I removed R59, R60 (47K), R2 and R29 (10K) forsome reason, probably because they added some pad function to thetransformer secondary that wasn't the right way to go. Sorry to befoggy on the details... but the mike line pad is still available andseparate from this pad.I bascially replaced all caps in the signal path (which are ceramicand mylar as stock) with polystyrene or polypropylene caps. One coulduse paper and oil caps if you want a more smeary 'vinatge' sound, butI didn't. The cathode bypass caps went from 47uF Al electrolytics to100uF solid Ta electrolytics. I hate recapping, and Kemet solid Tacaps should last a lot longer. As I recall, I replaced all Alelectrolytics in the signal path with Kemet solid Ta electrolytics,usualy of a slightly larger size. The output electrolytic couplingcaps were removed and replaced with jumpers.One important cap to remove is C6 and C16 (120pF ceramic). This capshunts the second stage plate and serves as an HF rolloff to make thecircuit stable. It is not needed with a lower ratio input transformerthat has no huge ultrasonic peak, and it only adds HF distortion.Clip it out, but only if you change transformers.The output stage uses a pair of dual op amps (BA4560 as stock), andthese should be replaced with something better. I have OPA2604 inthere now, but I might try the LT1469 because they have a cleaner HF.Not that this is a clean amplifier, mind you, but just to see if itsounds any nicer.The power supply can also be improved. The heaters run off ofregulated DC, so there's no issue there, but the HV supply can beimproved. As stock, it's a simple RC filter following the rectifier,and if you change this to a two stage filter by adding another filtercap and a resistor, you can get less noise and higher plate voltage.Reducing the value of the series resistor increases the resultingplate voltage, but worsens the filtering. Having two stages offiltering redcues the hum, but you end up with less voltage drop and ahigher resulting supply voltage, and that generally reduces thedistortion of the tube stage.Finally, the stock cheapo 12AX7 can be replaced with something better.I had a nice pair of Telefunken 12AX7, so in they went.When I have the time, I want to replace the output pot, as it seems tobe a linear taper pot or a really poor log taper pot. I want to geteven less gain from the tubes, so a change in the gain structure wouldprobably be in order. I might even change it to a single tube stagejust to avoid having to do so much feedback. A different tube typemight also be the answer there - something like a 12AT7 and new plate/ cathode resistors might be the ticket....

I will work on it tonight and post again tommorow

Ray

[/i]

Yeah, I realized the "y" is probably as valid a visual approximation of the Greek mu character as "u". I just couldn't resist. :wink:

FWIW, it seems you can get the mu character if you enter the HTML character entity "μ" in the message body. (eg. "2.2 μF" will render as "2.2 μF" )

I personally can't wait for PCs to grow to 80-bits so I can address a yottabyte of RAM (or would that be a yobibyte?)

Cheers!
--Bob

This is a difficult subject and can easily walk into audiophool land, that said I do find a sonic difference between capacitors. Not every time though and not in every app.

It's very hard to guideline since it depends a whole lot on the circuit, bias and so forth. No cap gives a warmer or phatter sound, imo. Some colour a lot some I can't hear. Also multiple caps do certainly ad up too.

Anything plastic, like polypropylene and polystyrene, tend to sound clear and hm... 'hifi'. Less clouded and more pleasing than regular electrolythic capacitors. Oil types mostly gives a dark and hmm.. 'euphonic', 'exciting' sound. That's probably due to some distortion but I can't worry 'bout that. Tantals seem 'grungy' to me, dirty, ballsy with a lot of presence. Ordinary electrolythics just tend to sound boring and clouding up things. Sounds 'electronic' to me..

There's no general guideline and I've heard very clear stuff with tantals and murky gear with polypropylene hifi caps. Oil caps always tend to put their trademark on things though.

And sometimes changing caps don't do jack, really.

I don't know where you've heard that tantals are the best in terms of tolerance and longevity, because I havehad almost the direct opposite experience :?

I don't know what to say about the SSL because I've never build one, nor am I likely to.

I played with one of those a year or so ago. I replaced the output opamps with Burr Browns (request them as samples), removed the ceramic cap that filters the high frequency to ground at the input (do this only if replacing input transformer with a 1:10 ratio part), replaced input transformer with a Jensen, replaced coupling caps with polypropylene's. I think the cap values were .047uF? Depending on when your pre was made it could already have larger value coupling caps installed. The original model used .022's I think. These cap values are from memory and they may be .47's and .22's? Check for oscillation at high gain if you replace the op amps. Oh ya replaced tubes with better 12ax7's.

NWSM