CD4049 vs CD4049UBE

[victorbignacio]

Hey everyone!
So I want to build RunoffGroove's Double D project.
Here is the RunoffGroove's perf board layout, it says the IC is a CD4049UBE: http://runoffgroove.com/doubled.html
But I will use the PCB layout from the official site, and it only says the chip is CD4049: http://www.runoffgroove.com/dd-wp-pcb.pdf

So... Is the PCB page just abbreviating the chip by ommiting the UBE indication, or the PCB layout use the ''un-UBE'' version?
What IC should I buy?
Thanks

[GibsonGM]

Welcome to the forum, Victor....I believe you NEED to have the unbuffered version of the chip for the circuit to work right...that is what "UB" stands for.

[Fancy Lime]

Hi Victor,

you can theoretically find all the information about that in the data sheet:
http://www.ti.com/lit/ds/symlink/cd4050b.pdf

Unfortunately, these are a bit confusing at first and need a bit of background info to understand.

The first important piece of information is, that for any type of overdrive/fuzz/distortion that uses hex inverters such as the CD4049 or CD4069, you always need the unbuffered version, never the buffered one. Only very few extreme noise fuzz circuits want the buffered version. The suffix "UB" means "UnBuffered", so you definitely want a version that has UB after the number, not just a B (which is the Buffered version).

The letter or letters after the UB denote the package type, packaging quantity etc. These chips come in many different packages. The "E" type is the "dual in-line package", the one that we use most of the time in DIY electronics. This is the package that fits the runoffgroove pcb. There are also single in-line packages, surface mount, small outline and many more.

So, long story short: You need a CD4049UBE, but if you buy it online it may also be called "CD4049 unbuffered dual in-line" or something along these lines. Do not take a buffered version and if you want to use the pcb make sure you don't accidentally get an SMD (surface mount device) package or anything of that sort.

Cheers,
Andy

[Mark Hammer]

Whether building a Double D, Tube Sound Fuzz, Red Llama, Hot Tubes, orany other invertor-based overdrive, always use the "unbuffered" version.

[victorbignacio]

Thanks to everyone! I will buy the CD4049UBE them.

[PRR]

20 years ago, 4049 was always unbuffered and did not say UBE.

Recently 4049 are buffered for better performance in logic circuits.

This spoils use in analog circuits.

So if you know the chip (or plan) is OLD, then it will be unbuffered.

If you have to buy new, for analog (audio) use, you now need to shop for UBE.

[snk]

HI,
Out of curiosity, if one was using a "non UBE" version of the 4049, would it result in the circuit not working, or not behaving as expected ?

[nocentelli]

distortion [linear, unbuffered] or square-waves [buffered, digital].

[blackieNYC]

The buffered one was used by Devi Ever in the Great Destroyer, which is a fun noisemaker. I like to think she had a bunch of the "wrong" 4049s and came up with something. 

[snk]

Thank you both !

[puzzlebunker]

The description of "buffered" and "unbuffered" variants doesn't make sense to me. Perhaps I'm missing something, but the 4049 is described in Texas Instruments datasheets etc as a buffer - so what on earth would be meant by a "buffered" buffer and an "unbuffered" buffer?

Whatever it means, it seems like the only thing you can get these days is the "UBE" variant. When I've tried suppliers who list 4049BE it either turns out to be unavailable or the chips that are shipped turn out to be 4049UBEs

[Kevin Mitchell]

What's so confusing about the title "unbuffered hex inverting buffer"

I know, I know...

There are two different uses of the word "buffer" here.
In the title "hex inverting buffer", it's for data conversion.
With the "buffered" variant, it implies additional circuity within the chip for current driving/sinking applications.

That's my understanding of it at least.

Anyways, what's wrong with using the unbuffered variant? Isn't that what folks are suggesting here?

[anotherjim]

The buffered inverters are basically each x3 inverters in series. They don't increase the output drive but are intended to speed up switching due to the increased gain. This can bite you if the input has a slow rise/fall time as the extra gain causes jittering which is a problem if it's clocking a flip-flop or counter. Then you should use a Schmitt trigger chip for slow sources such as an opamp as a comparator.
At any rate, the extra stages in a buffered inverter prevent it from biasing into linear mode as the unbuffered ones do which are single complementary pair per inverter.

[PRR]

.... the 4049 is described in Texas Instruments datasheets etc as a buffer - so what on earth would be meant by a "buffered" buffer and an "unbuffered" buffer?...

Lost in the mists of history.

The 4049 is not a great buffer, but a very handy thing for the price and the time (circa 1970). Note the "soft" transfer curve.

The 4050 is internally two 4049 in cascade. It is much less soft.

CMOS used to cost something. Pennies per gate. Today the cost has been driven-down SO far that some fool decided to slip in a buffered-buffered "improved 4049" under the same part number. A modern 4049 is even sharper than the old 4050. Great for decisive logic. Way too "hot" for analog audio signals. When complaints got knee-high they added some letter-code which seems to be unique in all of IC numbering.

[danfrank]

Here, these are buffered 4049 buffers. Good for the devi ever noise maker:

https://www.westfloridacomponents.com/IC583PE11/MC14049BCP+Hex+Buffer+IC+Motorola%C2%AE.html

These were half the listed price until I bought 30 of them. Then they raised the price, thinking they had something! Lol! They're gonna sit on the rest they have till the end of time!

Gotta love capitalism

[Rob Strand]

Mototola/ON also produce the Unbuffered type

https://www.alldatasheet.com/view.jsp?Searchword=MC14049UBCP


Mfg                 Datasheet        Pdip-16 part no.      Comment in datasheet

mot/on            MC14049UB    MC14049UBCP       "Meets JEDEC UB Specifications"
mot/on            MC14049B       MC14049BCP         "Meets JEDEC B Specifications"
Harris              CD4049UB      CD4049UBE
Natsemi          CD4049UBC    CD4049UBC
Toshiba          TC4049B          TC4049BP

xxx4049UByyy   = Unbuffered.     The important part is the UB.
xxx4049Byyy     = Buffered.          The important part is the B.

The yyy part is manufacturer dependent.  It usually means the type of package.
Different manufacturers use different codes for the same package.

You will find some parts are always B.
eg. CD4050B because by nature that have two stages inside.   
eg. CD4046B  because complicated chips cannot enforce UB.

If you can find a 4049B datasheet which has an internal diagram it will show three stages (as Jim already mentioned).

As for CD4049 I'd say that is often a generic part no used only simplified parts lists in catalogs.   At the end of the day it's ambiguous to whether it's U or UB.  It might not even be CD when you get it.

As for true "CD4049" parts, I actually didn't find a datasheet from natsemi or harris with CD4049B. (I did find an odd-ball part from another manufacturer.) I'll let one of you guys hunt it down - if it exists at all.

I'm sticking this on from the other thread since it does spell out some intent as to which devices are buffered and unbuffered, see p3,

https://www.ti.com/lit/an/scha004/scha004.pdf