TL071/LF351 VS NE5534an for amplifier/buffer

[darron]

hey guys. i'm at the stage where i'm getting itchy and wanting to design things. i think it would be a really good exercise for me to play with some opamps more.

first, i'd like to make a boost/buffer opamp pedal to call my own design. one knob. forums and articles seem very happy with throwing in a tl071, but what about the ultra low noise ne5534an? the tl071 has it's own jfet buffer and the ne5534an doesn't, so i'd want to put a jfet in front of it. i've read that an external jfet is usually better also. i've got some j201 jfets lying around. the aim would be the lowest noise, with best clarity and tone true to the instrument without flavouring. buffering with a j201 would drop a bit of volume wouldn't it? by the time that i use that with a ne5534an would it be just as noisy as the tl071 (which boasts low noise specs already)?

if anyone has any comments or anecdotes for me to consider then that would be really good (:

thanks in advance.

darron

[PerroGrande]

Both the 071 and 5534 are fine op amps, and I have designed successful circuits with both.

You've identified up front the biggest issue when dealing with a 5534 -- input impedance.  Obviously, it isn't an issue if you're using it in later stages.  However, if you're going for "stage one" boost/buffering (i.e. connected as the first thing after the guitar), it is something you have to think about.

Looking at the two Op Amps side by side for a moment:

TL071 has a higher noise figure (18nV/Hz^1/2  vs 4nV/Hz^1/2)
Both have the same slew rate (13V/microsecond)
The TL071 requires less supply power
The 5534 has better typical CMRR
The 5534 has better unity gain bandwidth
...

Welcome to the wonderful world of design trade-offs. 

If you're certain you want to go in the op-amp direction, I'd encourage you to research other chips, too.  For example, Analog devices makes practically as many flavors of op amps as there are stars in the sky... 

If you decide to go with an op-amp that will need a buffer in front of it, a J201 acting as a simple source-follower is always a winner.  It *does* produce a slight amount of *voltage* attenuation.  However, with proper design techniques and careful selection of components, the noise it introduces can be controlled. 

Also -- consider the various discrete options that are out there.  Op-Amps, especially when running in a single-supply world (as we frequently do when making effects) have some issues that need to be considered.

But above all, experiment!  Find out what works for *your* ears.  What I like may not be what you like, and vice versa.

[Gus]

It will be hard to make an opamp booster that has not been done before or in a book somewere or a circuit fragment in EDN etc.

But don't let that stop you from building and learning.  Buy some books the Walt Jung opamp ones are a good place to start.

now for you question some hints

Read about input pairs and how to design for low noise with the different types Jfet BJT.

Learn what the specs mean

Do tests with the sound change with input loading of different types of pickups.

[Mark Hammer]

Take a look at the op-amp comparison here in the Electronic Music Circuit Design article: http://hammer.ampage.org/files/Device1-8.PDF

[darron]

thanks for your tips everyone. sorry i didn't reply sooner, i was asleep with the time difference probably. i'll lean towards a ne5534an with a j201 infront of it, mostly in the style of the mxr microamp. that sounds like it would give me a bit more experience. i have been reading around heaps for the last few years. as for books, i've read craig anderton's and brian wampler's. i'd especially want the one that's better with unity gain.

so PerroGrande, to make up for the j201 attenuation, do i simply boost the opamp gain a little tiny bit from unit (about 1.1 gain)? and by proper parts selection, do you just mean keeping resistors values low and using metal films and avoid ceramics etc.?

I'll give that pdf a read Mark Hammer. Thanks. Thanks all really (:

[antojado]

Take a look at the op-amp comparison here in the Electronic Music Circuit Design article: http://hammer.ampage.org/files/Device1-8.PDF

That article by Anderton on opamps is really cool!

[Ben N]

i've read that an external jfet is usually better also.

Why? 

[darron]

i've read that an external jfet is usually better also.

Why? 

old thread. i can't remember who was saying it, but they said that a proper external jfet had a better buffering 'sound' to what the put in the op-amp packages. no matter, i've played around and found the OPA2134PA (and single package version) to be really nice. doesn't add any flavour or tone.

[George Giblet]

If you look a the J201 datasheet you will see the noise is higher than that of the NE5534.   True Low noise designs use low noise fets.  If you buy a modern low noise JFET input opamp it already has the low noise JFETs in it.   Why go through all the complication of a two stage design when you can just buy an awesome JFET input opamp off the shelf!

The other issue is when you design the gain stage you have to choose low feedback resistances, otherwise they will generate more noise than  anything else.

The guitar itself generates noise (I means hiss here not buzz etc) because of the DC resistance of the pickups. You cannot remove this hiss.  The inductance of the pickup makes things worse.  On a good day the pickup might look like a 10k resistor which equates to about 13nV/rtHz.

[brett]

Hi
there's a simple reason why the NE device on its own isn't desirable.  The input impedance is only 100kohms - about half of what I think is needed in a buffer (and many people want 1M !).  If you need a JFET buffer to go with the NE5534, you might as well use the TL071, I figure.

However, oddly enough, NE devices on their own are sometimes said to sound good.  Could this be a result of rolling off some highs?  Would Jimi Hendrix's solos have been as powerful in the midrange if he wasn't using those curly, treble-eating cables?
cheers

[mac]

My only experience with the LF351 is the Red Fuzz, I'm very pleased with it.

mac

[darron]

this is a pretty old thread and i steared away from the jfet. i didn't like the small loss of volume in buffering etc.

i suggest people play around with the OPA2134PA (:

[antojado]

Yes, I dug this up because of the Anderton article posted by Mark Hammer which has some really good info in it. I didn't want to start a whole new thread for it. 

[Mark Hammer]

It's a pity that the article was written in 1979.  It would be nice to see a similar comparison of many of the more popular/available op-amps that have emerged since then.

[frank]

I was searching for some opamps to play with so I looked at some schematics and took some notes.  I ended up buying:
lm741
lm308
tl071
tl072
ne5534
don't know what I will do with them, but it's a try.

My experience with opamps was only in industrial instrumentation so...

I fell on this "topic" right now and I am happy to see that those opamps are often used.

I bought a book on opamps a couple of years ago for 5$ in a junk store in the "nerd" section: Op Amps and linear integrated circuits (Boyce), I tought it might be useful for instrumentation...  Well it will be for an other purpose.

Small lineup of opamps:
http://www.dself.dsl.pipex.com/ampins/webbop/opamp.htm#select

If you have a lot of time to kill:
http://recforums.prosoundweb.com/pdf.php?th=5181&0/

[drewl]

I have a bunch of 5534's and they seem to be a bit quieter will a little less gain than 4558's.
Nice smooth breakup when used in distortions.

[darron]

I have a bunch of 5534's and they seem to be a bit quieter will a little less gain than 4558's.
Nice smooth breakup when used in distortions.

they should all produce pretty much the exact same amount of gain in the same setup. do you mean distortion instead?

[R.G.]

Welcome to design, as opposed to easter-egging parts or Frankenstein-ing (sewing together parts of multiple corpses.  )

To me, "design" always meant that the designer (a) knew a desirable result or destination, (b) knew at least the possible paths to that result or destination, and (c) had the ability to make intelligent choices (i.e. for a specific, stated reason) at points along the path where this became necessary, and (d) could state the reasons for making the choices along the path and the side effects/consequences of the choices.

As you're finding out, designing even a trivial opamp buffer is not all that simple. First you have to know the obvious, what opamp circuit does what, but hidden behind that is the need to know what the various imperfections of the real (not ideal) opamps that you can actually obtain will do to the circuit's performance, and what the various imperfections will do to the resulting circuit as compared to the desired result - whether the imperfections will matter at all, or make it unusable. If you're really good, you can project what imperfections will result in the opamp being audibly better in some way - and be able to back that up with fact, not advertising-speak if you're going to sell the box that results.

I have always viewed a formal electronics degree as being useful for two purposes. The minor one is instruction on how to cope with all this stuff in an organized way. The major one is weeding out the people who hadn't realized what was really involved in doing it right.

A formal degree is not at all necessary to do good design work. But a whale of a lot of work getting ready to design well is necessary.

[frank_p]

Mr. Rg, I was reading your comments on opamps yesterday and it reminded me what I used opamps in the first place: instrumentation, calculations and control.  It's easy to forget what it was used for in the first place when you are stuck in audio craze.

http://www.diystompboxes.com/smfforum/index.php?topic=53710.msg408423#msg408423
http://www.diystompboxes.com/smfforum/index.php?topic=53710.msg408486#msg408486

I don't see why beginners in audio (like me) would pay top $$$ for opamps if they don't understand the context of where the amp will be used.

Opamps are often viewed as a quasi ideal building block and a lot of the folks try to find what will be THE MOST PERFECT brick that will yield supranormal audio experience.  I think I will buy "jellybeans" for a long time before I understand all the design aspects of opamps.

It's not for nothing that I have posted these two links:

Small lineup of opamps:
http://www.dself.dsl.pipex.com/ampins/webbop/opamp.htm#select

If you have a lot of time to kill:
http://recforums.prosoundweb.com/pdf.php?th=5181&0/

Even for qualified people opamps versions seems to be a fighting topic.

What is great for understanding is small labs to breadboard with standard jellybeans to see what is the difference between them.  I mean the effects of their respective limiting factors (ex: slew rate, bandwith, loop gains, loading, etc....).  Comparing the differences to an other std opamp (741 per ex. since it is an historical ref. point) can give more understanding in the opamp world. 

It is sure that school force you to do some things that you would have not done by yourself and gives you a big edge.  Crunching numbers from 8H AM to 2H AM gives you knowledge (and nausea).