Hi there, I'm wondering if I could get some advice on changing the following circuit from 9volts to 18volts- advice on generally how difficult it would be, components that would need changing, etc.
I posted on this forum years ago, and with the help I got here, built and have been using the following circuit ever since: https://www.diyaudio.com/forums/ins...film-buffer-amp-combo-guitar-electronics.html
But recently I've upgraded my piezo film sensors to GraphTech ghost pickups. Overall it's a great upgrade, but now I'm getting a bit more of a "quacky" piezo sound. So the main intent here is to get rid of that, with the idea being that upgrading to 18volts will provide more "headroom" which will bring down the quack. I'm open to ideas if anyone has anything to say about this whole assumption.
I've attached the circuit below:
I posted on this forum years ago, and with the help I got here, built and have been using the following circuit ever since: https://www.diyaudio.com/forums/ins...film-buffer-amp-combo-guitar-electronics.html
But recently I've upgraded my piezo film sensors to GraphTech ghost pickups. Overall it's a great upgrade, but now I'm getting a bit more of a "quacky" piezo sound. So the main intent here is to get rid of that, with the idea being that upgrading to 18volts will provide more "headroom" which will bring down the quack. I'm open to ideas if anyone has anything to say about this whole assumption.
I've attached the circuit below:
I'm going to say nothing... nothing needs changing as such. Just check that any electrolytics on the supply (C4 and C5) are suitably rated. 25 volt for C4 and 16 or 25v for C5.
Also check all other electrolytic caps (all those coupling caps) are OK voltage wise. The 10uF caps see half the supply voltage, the 1uF could see a bit more or a bit less. If in doubt just measure the voltage across them and check its OK.
Also check all other electrolytic caps (all those coupling caps) are OK voltage wise. The 10uF caps see half the supply voltage, the 1uF could see a bit more or a bit less. If in doubt just measure the voltage across them and check its OK.
Awesome, thanks Mooly, for your insight, much appreciated. With this info I'll test the circuit and report back here with results. Thanks again!
Thanks for all the input/comments. I checked all my components for ratings, all was good, so I tested the circuit at 18 V yesterday. To my ears, the sound is very similar, I couldn’t really tell a difference between the two. This is actually good on one level, since that means I can keep using a single 9V and not have to worry about routing out a new battery cavity.
My current theory of why these new piezos sound harsher/quackier/less-natural, is that they are situated directly under the strings inside the saddles, whereas my old piezo film sensors were sandwiched between the saddle bolts and the guitar body (thereby picking up some resonances of the wood, and less direct sound from the strings). I’ve also been looking into and testing IRs, some of which sound promising, so this is probably the direction I’ll head in to try and solve this.
I actually forgot to check the LED when I tested this (it’s a push button “vandal” switch near the input jack on the bottom of the guitar)... I’m guessing it just wasn’t as dim in the 18V setup. I’ve set the LED very dim on purpose, just enough that I can tell if it’s still on when I’m done playing the guitar and lay it down or hang it up on the wall. Also, I’m assuming that the dimmer the LED, the longer the battery will last.
My current theory of why these new piezos sound harsher/quackier/less-natural, is that they are situated directly under the strings inside the saddles, whereas my old piezo film sensors were sandwiched between the saddle bolts and the guitar body (thereby picking up some resonances of the wood, and less direct sound from the strings). I’ve also been looking into and testing IRs, some of which sound promising, so this is probably the direction I’ll head in to try and solve this.
I actually forgot to check the LED when I tested this (it’s a push button “vandal” switch near the input jack on the bottom of the guitar)... I’m guessing it just wasn’t as dim in the 18V setup. I’ve set the LED very dim on purpose, just enough that I can tell if it’s still on when I’m done playing the guitar and lay it down or hang it up on the wall. Also, I’m assuming that the dimmer the LED, the longer the battery will last.
Last edited:
Modern LED's can be very bright with very low current... but not quite that low 😉
If you really want to extend battery life then you should look at possible alternatives to the OPA opamp and look at low current devices that also have good output voltage swing.
8ma draw for the opamp (which would be typical) gives about 50 to 70 hours running from an alkaline PP3. You should be able to find something that draws a lot less than that.
I wonder what the FET stages draw?
Have you measured the total current draw?
If you really want to extend battery life then you should look at possible alternatives to the OPA opamp and look at low current devices that also have good output voltage swing.
8ma draw for the opamp (which would be typical) gives about 50 to 70 hours running from an alkaline PP3. You should be able to find something that draws a lot less than that.
I wonder what the FET stages draw?
Have you measured the total current draw?
Thanks 🙂
So there is quite a bit of scope for cutting the current draw, perhaps increasing battery life by as much as 400% with it idling, maybe even more.
So there is quite a bit of scope for cutting the current draw, perhaps increasing battery life by as much as 400% with it idling, maybe even more.
Wow yeah that's a lot of battery life saved! Would the sound quality have to suffer for using a more efficient (if that's the right way to describe it) opamp? Or is mine a case of "overkill"?
I haven't measured the total current draw, but I will try and do that over the next week. I'll create a breadboard clone so I can test easily. Also to test other opamps.
I haven't measured the total current draw, but I will try and do that over the next week. I'll create a breadboard clone so I can test easily. Also to test other opamps.
As always, trust what you hear 🙂
Why not fit a socket and then you can try a few different devices. An ancient device that costs peanuts is the TL062 which is the lower power version of the TL072/82 series. Current draw is about 0.4 milliamps for the package, that give a battery life for the opamp alone of about 1250 hour
Here is the data sheet. These sound pretty good as well (you must listen for yourself and decide yourself though).
https://www.st.com/resource/en/datasheet/tl062.pdf
Can we do better? We can but the above cost buttons and you might just be surprised... give one a try...
There are some decent CMOS type opamps available now such as the TLC252. If you really wanted to explore the possibilities then I'd recommend starting a new thread asking for recommends for low power dual opamps.
Why not fit a socket and then you can try a few different devices. An ancient device that costs peanuts is the TL062 which is the lower power version of the TL072/82 series. Current draw is about 0.4 milliamps for the package, that give a battery life for the opamp alone of about 1250 hour
Here is the data sheet. These sound pretty good as well (you must listen for yourself and decide yourself though).
https://www.st.com/resource/en/datasheet/tl062.pdf
Can we do better? We can but the above cost buttons and you might just be surprised... give one a try...
There are some decent CMOS type opamps available now such as the TLC252. If you really wanted to explore the possibilities then I'd recommend starting a new thread asking for recommends for low power dual opamps.
Ok I'll do that... fit a socket and so I can test out the TL062, and maybe others. Thanks again! I'll post results when I get them.
sounds like a plan.Another good reason for the planned approach is, a breadboard of the same circuit will only give representative results if the J201 FETs closely match the ones in the circuit in Idss, which is VERY unlikely. Often a JFET may have an Idss spec that varies over a range of 3:1 or 4:1, to as high as 6: or even 8:1. The J201 is no exception: The 2015 PDF from OnSemi/Fairchild shows an Idss of 0,2mA to 1,0mA @ Vds=20 a 5:1 ratio. Your 6-FET buffer could use as little as 0,6mA, or as much as 3mA(*), depending on fitted FETs -- and all of them would be within spec!
* the actual values will be a little lower, since the Vds will be 3,5V to 4,5V
Glad you're off of the supply-doubling idea -- it solves fewer problems than one might think.
Cheers
* the actual values will be a little lower, since the Vds will be 3,5V to 4,5V
Glad you're off of the supply-doubling idea -- it solves fewer problems than one might think.
Cheers
Last edited:
Thanks Rick for the info about the J201's... I recall it just slightly from before, but great to be reminded of this. I'll measure and note the Idss of each one individually in the future. I'll try out a few in the breadboard with as varying Idss as I've got... and see how that affects the sound.
You're welcome! 😀
But umm-m -- I only mentioned it because it affects the battery drain .. the sound quality should be unaffected -- unless you overlook the recommendation about the Idss's of the FET pairs: Q1 >> Q2. The Source follower transistor, Q1, must be 'stronger' than the sinking CCS, Q2. Otherwise the CCS will pull down hard enough to partially forward bias the Source follower.
And of course, it applies to each of the three pairs.
Regards
But umm-m -- I only mentioned it because it affects the battery drain .. the sound quality should be unaffected -- unless you overlook the recommendation about the Idss's of the FET pairs: Q1 >> Q2. The Source follower transistor, Q1, must be 'stronger' than the sinking CCS, Q2. Otherwise the CCS will pull down hard enough to partially forward bias the Source follower.
And of course, it applies to each of the three pairs.
Regards
Ah ok, yeah I knew about the forward biasing, and to measure each one to determine its place, but I did not know that the overall Idss's affect battery drain differently, so thanks! Another piece of the massive puzzle. I'll keep track of the Idss's going forward... I don't know if I'll be able to determine the correct Idss's of the FETs in my already soldered circuit though... the other components in the circuit might affect that? I'll see if I get any reasonable values.
Last edited:
Just measure the current draw with opamp removed. Most of what is left will be down to FET current draw... assuming you actually have that high value R in series with the LED 🙂
Since the presentation of the infamous ovation guitars I dislike piezo bridge pickups for their harsh sound. You can hear that annoying sound on many acoustic guitar recordings performed by Eric Clapton. I did many experiments with these, but gave up at the end.
My explanation is that in bridge position with its node of oscillation excursion the first attack of the guitar pick or finger nails becomes dominant giving an additional percussion peak causing that unnatural sound.
My explanation is that in bridge position with its node of oscillation excursion the first attack of the guitar pick or finger nails becomes dominant giving an additional percussion peak causing that unnatural sound.
Not that you want to reinvent the wheel, but With a pair of Double-A's and a MAX232 chip you can get +/- 10V. Other newer "charge-pump" chips are more tolerant of high input voltages. While I've seen folks discussing this for low power audio, never implemented it myself.