Psst... Ready to join TalkBass and start posting, make new friends, sell your gear, and more?  Register your free account in 30 seconds.

Opamp Supply Voltage

Discussion in 'Pickups & Electronics [BG]' started by daveze, Dec 7, 2004.


  1. This probly isn't strictly a pickup question but its kinda related to the 9v vs 18v sorta thing a bit. I've done some research and I've found out what opamp is running inside my Yammaha BB-G5A, which is the RC4558. Its supply voltage is ±18V but is currently running off a 9V, I can't see whats happening between the battery and the IC but I'd have a stab and say nothing, so it'd be recieving close to ±9V.

    What I'm wondering is what happens to an opamp when you exceed the supply voltage? I've had the 18V conversion in the back of my mind for a fair while but I haven't been sure about what'll happen if the supply voltage of the opamp is exceeded.

    Josh D
     
  2. Passinwind

    Passinwind I Know Nothing Supporting Member

    It's probably floating the op amp input at 4.5 volts, and seeing +9 and -0 on the V+ and V- pins. Grab a meter and find out, it's pretty easy to check. In which case, two 9V batteries in series would put you at 18 and 0,with the input floating at 9, more or less, and you'd still be way below the point of worrying about it in terms of the opamp, but the bypass and signal caps need to be rated high enough to deal with the increased voltage as well. PM me if this isn't clear enough, OK? I can walk you through all the things you'll need to check.
     
  3. RyanHelms

    RyanHelms

    Sep 20, 2003
    Cleveland, OH
    Hey Passinwind, would it be likely that the caps are already 25v or better, or would they have put 16v or less jobbies in there?

    Here's the data sheet for the IC.
     
  4. I'd say it's highly unlikely that they use anything under 50v caps. The only case where it becomes important is when you need bigger electrolytic caps, like in a power supply, where voltage rating and capacitance dictate the size. Check the bigger caps in there, sometimes you can see the voltage ratings on them. Of course, if it's epoxy dipped like a J-Retro, your out of luck.
     
  5. chucko58

    chucko58

    Jan 17, 2002
    Silicon Valley, CA, USA
    I paid for all my gear myself. Well, me and MasterCard.
    A 4558 ought to be good up to 30 V (+/- 15 V) so 18 V isn't going to hurt it a bit. In fact, a 9 V battery is a little marginal for these chips, especially once it's drained down a bit.
     
  6. Passinwind

    Passinwind I Know Nothing Supporting Member

    You never know, especially if it has bigger value electrolytics on the power supply rails. I spend a lot of time cleaning up messes folks made by assuming things are safely overrated. :eyebrow:

    Honestly, I haven't worked much on stock onboard preamps. I like to roll my own, and I use external power, and lots of it. RC4558 wouldn't be high on my list of choices, but if it sounds good, it is good, I guess. :cool:
     
  7. xyllion

    xyllion Commercial User

    Jan 14, 2003
    San Jose, CA, USA
    Owner, Looperlative Audio Products
    Although it is possible, it is unlikely that any capacitor goes across the full power supply voltage. They need to develop a reference voltage that is half of the supply voltage. Electrolytic caps are often used to stabilize this reference voltage. Other electrolytic caps may be used as DC blockers on the input signal, but those won't be affected by the supply voltage.

    Although I couldn't say for sure without examining the preamp, it is most likely ok for you to run the preamp at 18V.
     
  8. They should be rated at 100V, thats the green caps, the others don't say but they're all like 1D35, 1B30, etc for what thats worth.

    So in theory I'd be able to go up to 3x9V? I won't be stuffed if I kill the sucker, I've got a OBP-3 thats collecting dust but it would be a waste.

    Passinwind, I too would like to make my own preamp too. Only problem is I haven't studied any electrical stuff since high school, so my understanding is limited, very limited. I don't completely understand why it sees +9V and -0V, its because the -ve connects to the ground, right?

    Why would the RC4558 not be high on your list, what are examples/characteristics of good opamps?

    Josh D
     
  9. RyanHelms

    RyanHelms

    Sep 20, 2003
    Cleveland, OH
    Right. And right too about 3x9v, that gives you +/-13.5v swing. Hey guys (passinwind and xyllion) would running that much closer to the rails improve the opamp performance? What's the difference between running at 27v or 18v or 9v ? Longer battery life? Increased headroom? Doeasn't the opamp perform the same so long as the +/- voltage swing is equal and doesn't exceed max?
     
  10. xyllion

    xyllion Commercial User

    Jan 14, 2003
    San Jose, CA, USA
    Owner, Looperlative Audio Products
    It all depends on what the voltage swing of the signal is. Let's say you have a signal that is 3Vp-p. That leaves 6V to play with on your 9 volt battery. Now your op amp is incapable of handling a signal that goes all the way to the edges of the power supply rails. So, now you give up another 2V (not sure what it actually is for the 4558). That leaves 4V of headroom. As the battery begins to die, that headroom gets smaller and smaller until finally it is all gone.

    With an 18V system, you need the same 5V for normal operation, but now you have 13V to spare. You will have more headroom for a longer time with the two batteries. The down side is that now you have to make room in your bass for a second battery. Also, when it does get to be time to change batteries, you will need to have 2 batteries available.

    The same argument holds true for 3, but the added benefit isn't worth it for most people. Having 3 batteries takes up quite a bit of space and even though you don't need to change batteries as often, when you do you will need to have 3 new ones in your bag.

    Personally, I prefer to pick op amps that can run closer to the rails with nice low noise margins. A 9V system can work just fine if your pickups aren't too hot and if keep the gain in the preamp at a reasonable level.

    With the preamp you are using, you probably will see some benefit from an 18V supply.

    Best of luck.
     
  11. Passinwind

    Passinwind I Know Nothing Supporting Member

    I agree with pretty much everything xyllion said above. The big benefit to increased voltage is often a better noise floor. I do have two basses and one guitar with 9V preamps, all of which work very well.
     
  12. chucko58

    chucko58

    Jan 17, 2002
    Silicon Valley, CA, USA
    I paid for all my gear myself. Well, me and MasterCard.
    IMHO the big benefit to increased supply voltage is more headroom. If that means you can run more gain in the on-board preamp, that could result in a lower noise floor, but it's by no means guaranteed.

    An on-board preamp powered by a single 9V battery could conceivably drive most power amps to their rated output. Yes, I mean you could plug the bass directly into the power amp if you were so inclined. How much more headroom do you want?
     
  13. I took a quick look at the specs of the 4558 and it looks like a decent opamp. According to Texas Instruments ( not a bass manuafacturer ;) ) the op amp really should have at least 10 V to operate at it's best. It also sucks 1v at each rail so you have 7 volts of signal swing. It takes 2.5 - 3 ma of current, which is about average to some other amps. I can't tell from the specs but it might have some crossover distortion which would add some distortion when playing at low levels. I checked around for some other recommended op amps, and they cross reference to the 4558. One could loook for a rail-to-rail op amp to give a little more headroom, or a lower noise op amp for a cleaner signal, but the cost-to-performance ratio is pretty good with this amp. I looked in my Peavey preamp and it has 4558s sprinkled liberally throughout. (Maybe that's a condemnation of this opamp?) The higher performance op amps generallly have higher current requirements, and need a higher voltage power supply to be at their best.

    BTW, I think it's Aguilar that uses a voltage doubler to get 18 volts out of a 9 volt battery. Pretty cool huh?:cool:
     
  14. Passinwind

    Passinwind I Know Nothing Supporting Member

    The 4558 is standard issue in many pieces of pro gear, and it works pretty well there. As others have mentioned, it's happier with a higher rail voltage than a 9V battery supplies. There are voltage doubler chips out there now; some even do a bipolar supply off a unipolar input, and for all I know, the Yamaha preamp might even use one. 4558 is part of the "mojo" of the Tube Screamer, it does have a characteristic distortion thang going on. I use NE5532 as a sub a lot of times, if there's sufficient margin in the power supply for the increase in current draw. I also prefer the very common TL072 in many cases, or better, TL062 for low power applications. I just like the distortion character (or lack thereof) of these better than 4558. But be aware, many other folks hate all of these choices passionately too.

    For really low distortion and noise something like OPA 2604 or OPA 2132 are well regarded. These draw a lot more current than many others though, and are probably not such a good choice for battery operation, same withthe 5532. They also drive low inpedance loads more happily than most, but need at least +/- 12V to perform optimally, IIRC. I just subbed a bunch of these into my Ashly bass preamp (which uses 15 RC4558s), with very nice results. However, a properly designed 4558 circuit works just fine in many situations, and substituting an opamp with a FET front end isn't always an improvement. It can be when fed with a high impedance source like a bass pickup though. When I'm building my own stuff, there's little reason not to use a $5 opamp rather than a 30 center. Since I can design around the chip design, I like to start with the best chip I can afford. Honestly though, diminishing returns can kick in very easily. :cool:

    A friend of mine designs and builds pickups. He just sent me a few LM6172s, which purport to be a good low power opamp. I'll report back next week, when I try them.
     
  15. Just checked the National Semi web site. The LM6172 looks like a great amp. The minimum supply voltage is 5.5v, so low batteries and such shouldn't be much of a problem. The slew rate is 3000v/us, about 1000x higher than the 4558. This is good for transient response. One negative is the internal noise is somewhat higher than the 4558. Depending on how much gain you need, that may or may not be a problem. Also the supply current is a little bit less than the 4558, meaning you'll have longer battery life.

    We used it at a company I used to work for and had good results with it for our application (computer audio) but it was too expensive to put in a consumer product. Instead of the $1.60 6172, we used a $0.25 LMV358. The LMV358 is a great product for low voltage, portable applications: it runs rail to rail, has low current requirements (0.1 ma!), and is cheap. The shortfall is that it only goes up to 5.5v supply and the high noise compared to the others.
     
  16. xyllion

    xyllion Commercial User

    Jan 14, 2003
    San Jose, CA, USA
    Owner, Looperlative Audio Products
    That is the unfortunate law of physics. You can't have everything all at once. Low cost, low noise, and low supply current all work against each other. As a result, you need to select an op amp that gives enough of each to make you happy. Unfortunately, not every preamp manufacturer takes the time to research what is on the market.

    I suspect the 4558 was chosen because they were familiar with it and not because it is the ultimate choice. I know that John East did research different op amps for the J Retro and the one he chose was not the 4558.
     
  17. chucko58

    chucko58

    Jan 17, 2002
    Silicon Valley, CA, USA
    I paid for all my gear myself. Well, me and MasterCard.
    The 4558 is a device that dates back to the early '80s. Its primary virtue is that it is very cheap. (The chip makers refer to generic, low-cost chips like this as "jellybeans", which should tell you something.) There are many better choices today.

    The chip that Carvin uses is the ON Semiconductor MC33178. It is a low-noise, low-power dual op amp that is capable of driving 600 ohm loads. It is also relatively inexpensive.

    The chip John East uses in his current J-Retro devices is similar to the MC33178, at least from the spec sheet point of view, but is more available in Great Britain.
     
  18. Just curious so I did a little economics study. Prices are from digikey in quantities of 2000 or more.

    National LM6172 $1.82
    Mot MC33178 $0.334
    TI RC4558 $0.119

    You could probably get them even cheaper if you were using a million of them a year. There are plenty of cheap clock radios, boom boxes, and other stuff like that to make it a realistic number.

    Let's take my Peavey Probass 1000 preamp, for instance. It has about 15 op amps in it. If it uses the 4558 (which it does), and you build 5000 every year, you need 75,000 opamps. at 11.9 cents you spend almost $9000 for those amps. If you upgrade to the Motorola, thinking "hey how much difference can 20 cents make?" then all the sudden you have invested over $25,000 in op amps, so you net profit has just fallen by about $16,000. And that's only one product. They probably use truckfulls of them across the product line.

    Now if you put the LM6172 in there you don't have to do the math to see why peavey or Carvin or Fender don't use anything but the cheapest op amps. Maybe Euphonic Audio or Walter Woods can do it because they charge more, and have a much better reputation for sound. But then again you pay for it. So building your own preamp will theoretically give you way better sound, and since you're only spending $2.00-$3.00 more for one op-amp, it's a no brainer.
     
  19. A9X

    A9X

    Dec 27, 2003
    Sinny, Oztraya
    Beware that the 6172 can be a real bitch to get stable and not oscillating. Performance will be very layout dependent, and in the relatively low gain you're going to need for an onboard pre, it's gonna have way too much bandwidth. Good supply decoupling is essential, including the bias network (I don't know how well it will work single rail either), it will need a series resistor to isolaite it from cable capacitance and probably a groundplane especially if you want to use it in an RF rich environment.
    It's performance isn't so great that it would be worth the effort for me to use it over some of the others suggested.
     
  20. Passinwind

    Passinwind I Know Nothing Supporting Member

    Heh, he emailed me this morning to suggest that I try it as a line driver, echoing many of your points. :cool:

    As far as onboard preamps I build, I do the outboard power thing anyway, so low current and single supply issues are moot for me.