What do 4558's sound like?

[shaddai]

The reason I ask is that this Tascam 32 I just got has a couple of JRC 4558's and 4559's as the main record/playback IC's. I also have an Otari MX5050 8SD that I really like the sound of. Both machines are cleaned and aligned to their respective MRL's, running 456. In fact, the Tascam is probably lined up more acurately than the Otari.

Overall, the Tascam is okay I guess, but if I were to give a 100 score to the Otari, the Tascam would get an 85...and my DAW would get a 98. To describe the sound of it versus the Otari I'd say it's: kinda flabby & sloppy on the bottom, mids are less detailed, and sorta harsh/edgy. The highs are almost digital sounding, and I know the 12k knob is much more fun with the Otari than the Tascam.

Also, it seems like it's just dynamically challenged. I thought I might have missed something and been hitting the tape pretty hard, so I dropped a couple of mixes I'm really familiar with at progressively lowering levels just to find out that I'm not crazy. It just rides the short bus when it comes to dynamics.

I do know people don't like the 4558 very much unless it's a guitar pedal. Is this what I'm hearing? Do you think an LM833/5532/OPA replacement would wake the deck up & silk it up some?

I'm already planning on going through & replacing the caps & resistors in the signal path with updated versions of the original values, but just wanted some guidance on whether I should change out those IC's.

Thanks
Todd

 

[gyraf]

The limitations are most probably in the tape handling and magnetics - don't expect to get much results from component upgrades. 4558 is by no means a bad opamp, if treated right.

Jakob E.

 

[cuelist]

Most general purpose opamps sound good when used within their limitations and in a properly designed and layed out circuit.

The problems occur when you are trying to feed them higher bandwidth signals than they are capable of handling, asking them to drive too low impedances (incl the f/b network) or asking individual opamps for silly amounts of gain (running out of GBW).

Like Jacob said, the limitations in your case are probably elsewhere.

I'd look into replacing and upsizing all electrolytics. I would not bother with replacing resistors or film caps. In the grand scheme of things, these have microscopic inpact, at best.

 

[NewYorkDave]

You've discovered the legendary "Tascam sound." :wink:

I'm usually very skeptical of the efficacy of chip swapping but in this case I think it might actually help, if done carefully. There's really nothing badly wrong with the Tascam heads and transport as such. Thing is, you really have to know what you're doing to attempt a chip upgrade; I cringe when I hear people say "replace all TL072 with XXXX" because a replacement that works great in one circuit might be disastrous in another. Tape decks are also tricky because of the record/repro EQ and the drive requirements of the record and erase heads. Upgrades should only be attempted by someone with the know-how and equipment to measure the effects and also to perform the necessary re-alignments after the components are replaced.

Short answer, if you're looking for an already worked-out solution, I'm afraid it doesn't exist. An upgrade is possible but it's not necessarily practical unless you're a hard-core experimenter.

 

[bcarso]

[quote author="NewYorkDave"]You've discovered the legendary "Tascam sound." :wink:

I'm usually very skeptical of the efficacy of chip swapping but in this case I think it might actually help, if done carefully. There's really nothing badly wrong with the Tascam heads and transport as such. Thing is, you really have to know what you're doing to attempt a chip upgrade; I cringe when I hear people say "replace all TL072 with XXXX" because a replacement that works great in one circuit might be disastrous in another. Tape decks are also tricky because of the record/repro EQ and the drive requirements of the record and erase heads. Upgrades should only be attempted by someone with the know-how and equipment to measure the effects and also to perform the necessary re-alignments after the components are replaced.

Short answer, if you're looking for an already worked-out solution, I'm afraid it doesn't exist. An upgrade is possible but it's not necessarily practical unless you're a hard-core experimenter.[/quote]

Concur. Many parameters enter in to opamp selection and blind substitution is rarely productive unless you love to putter and measure and listen. Slew rate, gain-bandwidth, input current noise, input voltage noise, open-loop output impedance, open-loop distortion, common-mode rejection and distortion, crossover distortion, open-loop output impedance, available output current, overload behavior....the list goes on.

 

[SonsOfThunder]

First order of business when considering OA upgrades:

Do you have a schematic?

I have to agree that much of the time, the BW limits are set by the surrounding circuit values rather than the opamps. (Yamaha MC series mixers...)

:grin:
Charlie

 

[shaddai]

Do you have a schematic?

Yup..and the 105 page tech manual with the parts list, and the tascam official description of "when you do this, this happens in the circuit". Tascam definitely knows how to write a tech manual. Anyway, the design seems pretty basic.

I'm usually very skeptical of the efficacy of chip swapping but in this case I think it might actually help, if done carefully. There's really nothing badly wrong with the Tascam heads and transport as such. Thing is, you really have to know what you're doing to attempt a chip upgrade; I cringe when I hear people say "replace all TL072 with XXXX" because a replacement that works great in one circuit might be disastrous in another.......

Well, I may still be a bit new, but I'm not 100% in the dark. I'm at the point now where I can formulate and understand what I want to do, but still go back and double check everything I can with the book & calculator before jumping in. I'm definitely not 'newbie' enough to think I could resolder a chip and a few caps and come up with a Studer. If that was the case, Tascam would have beat me to it.

If it takes a month of 'hobby time' and pokin around with the scope to get a solid improvement nailed down, then that's fine...that's how I learn the best anyway.

I suppose for now, I'll work with cuelist's suggestion and look over the caps some more.

Thanks for the help. If I come up with anything dramatic, I'll let you guys know :thumb:

Todd

 

[SonsOfThunder]

[quote author="shaddai"]Yup..and the 105 page tech manual with the parts list[/quote]Scan the relevant part(s) of the schemo and post it here if you'd like some commentary.

Well, I may still be a bit new, but I'm not 100% in the dark.

Awesome, you ought to know by now that you fit in here just fine with the rest of us.

If that was the case, Tascam would have beat me to it.

Maybe not...cost restraint is often a performance killer. Granted, you aren't gonna get Studer performance, but you might get a pretty killer machine above what you have now.

If it takes a month of 'hobby time' and pokin around

That's how I learn every day. I just happen to have a job as a "scope poker" (actually PRR labelled us as solder pokers a while back :grin

If I come up with anything dramatic

We could have a *discussion* here! :razz:

Peace!
Charlie

 

[bcarso]

[quote author="RogerFoote"]Personally, I wouldn't be afraid at all of swapping out op amps as long as you don't overload the psu... Just download the datasheets, do some comparative analysis and go. Solder in a socket where the old op amp is, then if it IS a "disaster", just plug in a 4558 or TL072 to put it back the way it was.[/quote]

But don't be about driving your tweeters until you've verified things aren't oscillating like a house on fire...

 

[Svart]

I can tell you from my puttering experience with opamps ranging from .02$ to 20.00$ a piece is that phillips/signetics 5532s with Nichicon (Muse) caps will put you smack in the center of "high end" without breaking your budget. everything else applies though, but these guys are tough and stable, at least moreso than the higher speed/bandwidth parts.

 

[SonsOfThunder]

[quote author="cuelist"]replacing and upsizing all electrolytics.[/quote]Do you mean upgrade (as in putting in low ESR caps instead of generic junk) or increase in value?

If you are increasing in value, by how much? I have gone by a rule of thumb taught to me by our sr. design engineer and made a lot of measurements afterward. The rule works out pretty well.

If you are talking unbiased electros in the signal path, I like to replace with equivalent size film or at least put a 0.1 or 1.0 film in parallel.

Hope that is helpful!
Charlie

 

[Svart]

The signetics/phillips 5532 is pretty good. I recently got a whole batch of them and populated a pair of channels in my console. i used nichicon KZ muse caps which I also got in a deal. My channels use 10 dual opamps and 2 quad opamps in the EQ section. for the channels with FET input opamps i used opa404 as the quad opamp. Because of the FET inputs, DC offset is very low which allows me to use NO coupling caps on these channels. trying to use NO coupling caps with BJT input opamps will likely render your channel useless due to insane DC offsets. there are ways around this but the coupling cap is the cheapest and easiest.

the mics used were m930s and mc012s, the preamps were 1272s to an MX2424 24bit 48k with opa2604/stargets on it's I/O.

In a pair of channels i used opa2604 with NO coupling caps.

In a pair of channels i used opa2134 with NO coupling caps.

in a pair of channels i used opa2604 with HE nichicon caps.

In a pair of channels i used opa2604 with PW nichicon caps.

in a pair of channels i used opa2134 with HE caps.

in a pair of channels i used opa2134 with PW caps.

in a pair of channels i used mc33078 and mc33079 with HE caps

in a pair of channels i used ne5532 and mc33079 with HE caps

in a pair of channels i used ne5532/mc33079 with KZ(muse) caps.

I also have channels with opa627/637 with the different caps as well as ad825s and the different caps but the high price and/or the need of adaptors for all of the parts causes the total cost to be extremely high, however i will tell you that the ad825 is a remarkable part and when used in summing areas can really make a difference. I use this part in my master channels until i can make enough adaptors to use it in all the buss/master areas. Of course i have also taken the time to adjust feedback for each part to stop oscillation if any. you will need to do this for your board too as many have mentioned.


no matter how you slice it the channels with NO caps were insanely detailed and crisp but almost unusable due to the brittle sound. cymbals sounded like they were made of glass when boosting a bell at 15k

the channels with HE caps were very detailed but dry, the 2604 having a hyped high mid, the 2134 having a dip in the high mids. boosting 15k gave a shrill sound even though it wasn't really harsh.

the PW caps are a good match for the FET opamps due to being slightly dark, but why buy 4$ a piece opamps just to use caps to negate their sound? boosting 15k sounds ok but nothing special.

the mc3307X channels sounded good but fairly neutral. none of the caps seemed to affect the sound much. boosting 15k gave no strangeness but still sounded dead to me

the 5532 channels all sounded similar but since i had a big bag of muse caps i'm just stuffing those from now on. the 5532 channels sounded less harsh than the FET channels almost sounding slightly dull BUT when mixed with other channels they seemed to retain their detail whereas you usually seem to lose some of the detail when summing. more detail in the mids was also noticed, less detail in the highs and lows. an interesting outcome. I was able to boost a bell peak at 15k without any harshness which caused the detail in the highs to come back.

i am in the process of working on the clips of this shootout(which i mentioned some time ago in another thread) which should be available in the next month or so.

the short answer is that even the expensive parts can be a waste of time and money in certain circumstances. My channels don't use a lot of opamps and even less so if i bypass things like the EQ so I'm not hearing a lot of difference. most of the opamps are in buffering configurations with only 2 in gain arrangements if the EQ is switched out. I plan on making adaptors for the FETbloaks to go here anyway.. :grin: For those people with channels that use TONS of parts the effect of different parts will be much greater.

 

[bcarso]

[quote author="RogerFoote"]Were those 5532s made before the famous fire?[/quote]
"We?re having a heat wave, A tropical heat wave, The temperature?s rising, It isn?t surprising, She certainly can Caen-can."

 

[Svart]

Bcarso... Drunk again!

The Signetics have this info on them

S
NE5532N
FPA3425
9340v1


the Phillips parts have this info on them:

Thailand Phillips
Ne5532N
0178A13
9740NJ

I never bothered to figure out years on them, I just let the ears figure it out. :green:

 

[shaddai]

The signetics/phillips 5532 is pretty good. I recently got a whole batch of them and populated a pair of channels in my console. I used nichicon KZ muse caps which I also got in a deal. My channels use 10 dual opamps and 2 quad opamps in .........(along with the rest of his post.)

What a fantastic post! You would not believe how much help that is! I'd built a quick and dirty headphone amp on the breadboard to see which chips I like the character of, but your post is opening my eyes to chips I hadn't though of yet. I haven't done any spec checking yet, so some of the chips I've checked out may very well not work in the circuit, but it is leading me to a general area of what I'm wanting. It's interesting you'd mentioned the AD825. My favorite in the pile so far is the AD845JN.

You'd mentioned adapters with the 825. Have you found an adapter that will allow you to use a single IC on 1/2 of a dual socket?

Thank you very much for your time!

Todd

P.S. Yes I have a scope

:grin:

 

[Svart]

http://cimarrontechnology.com/

they make nice adaptors. I make my own though.

Since i assume you want through hole parts and not SMD parts(the ad825 is SMD only..) you need adaptors.

Problem is that the ICs are 4$ a piece and then you buy another 4$ adaptor. that 12$ for a dual opamp. the opa627 is 15-20$ a piece and then you need another 4$ adaptor, that's possibly 45$ for a dual opamp. this may be worth it in summing areas that use a few opamps but need opamps with speed. for console channels that don't need warpspeed opamps but need a lot of good opamps the ne5532 is good enough if used properly.

EDIT: I use the ad825 because it's one of the few AD parts that can stand +/- 18v rails(my standard), most of the others are only rated for +/-15v.

There is tons of information that goes into much more depth if you search for it.

:thumb:

EDIT: RogerFoote, when i can compile the rest of my data and audio clips into a post I'll ask the admins to make a meta for comparing parts, however it is VERY subjective. I was thinking more along the lines of creating some PCBs with nothing but 20 opamps in a unity gain arrangement in series with each other with coupling caps between them. this way we can measure and compare the audio between each opamp and before/after the whole line. we can trade out parts or install jumpers to change our testing. this should satisfy all the skeptics and once posted, anyone could make the PCB and test for themselves. it would be simple to do A/B studies this way and also post some real numbers. However, some opamps sound different when loaded, some sound different when powered differently also. this should be taken into consideration too.

 

[shaddai]

Thanks for the link...just cost me $12 :green:

 

[evm1024]

I've had OPA2132 in mt Tascam 32 for the last year. I found it to be a significant improvement.

Will the "sound" work for you? Who knows. I primarally record classical pipe organ sterio direct to tape.

 

[Minion]

Well My Crappy Guitar amp used a 4558 Opamp and this amp sounded awefull so yesterday I swaped out the 4558 and installed an IC Socket and Put in a OPA2107 and the Change is Sound Quality was amazeing, Like Night and Day...Before it has No sustain and No mid or High end and just a very Muddy Low end and the Gain was really Bad sounding with the 4558 but after installing the OPA2107 (Burr Brown) the Guitar has a Lot of Nice Mid and High end and it has a Very Clear and Crisp sound with LOTS of sustain.....There is still a Bad Backround Hum and the amp still sounds Cheap but it is still a Major Improvement over what it sounded like before.....

So I think if you swapped out your 4558"s to some better opamps you might get a Better sound ......

Cheers

 

[StephenGiles]

I would say leave well alone. In the time you spend looking at it, thinkng about it, going to the electronics store to buy parts, possibly cocking up the unit....you could have recorded some stuff - which would be fun, and I doubt very much would sound much different to your ears than with xxxxx opamps.