Building A 4558 opamp from scratch.
hello guys, someone planted a bug in my head saying that guitar pedals sounds amazingly better with no opamps or chips in them and so I desided to try and build
a tube screamer with the opamp cirtuit included in the pedal circuit.
in other words I'm traying to build a 4558D opamp from scratch, now |I know that may sound very stupid, aspacialy when you can get them for less then a dollar
and it will take a lot of space and make the circuit a lot bigger.
I'm doing it from pure curiosity.
so i found the sircuit schematic and even pages that explain things but nothing about the values of the parts or the transistors types.
can anyone help me figure it out?
did anyone try this before ?
a tube screamer with the opamp cirtuit included in the pedal circuit.
in other words I'm traying to build a 4558D opamp from scratch, now |I know that may sound very stupid, aspacialy when you can get them for less then a dollar
and it will take a lot of space and make the circuit a lot bigger.
I'm doing it from pure curiosity.
so i found the sircuit schematic and even pages that explain things but nothing about the values of the parts or the transistors types.
can anyone help me figure it out?
did anyone try this before ?
- IvIark
- Tube Twister
Information
I've seen someone do this before but I can't for the life of me remember where, but I remember seeing a pic of a big PCB with the opamp circuit in conventional through hole components. We've probably all seen schematics for the ICs, and the 4558 isn't as complicated as some of them, but as far as values go I have no clue and unless a more complete circuit document can be found from JRC or TI (or whoever else has manufactured them) then you're in for a hell of a lot of trial and error.
As a pointer on the transistors, you can see they're all BJTs with the exception of the output which is a FET, and you can tell whether the BJTs are NPN or PNP by the direction of the emitter arrow. Apart from that you're on your own with your breadboard but I will be interested in hearing how it turns out.
As a pointer on the transistors, you can see they're all BJTs with the exception of the output which is a FET, and you can tell whether the BJTs are NPN or PNP by the direction of the emitter arrow. Apart from that you're on your own with your breadboard but I will be interested in hearing how it turns out.
"If anyone is a 'genius' for putting jacks in such a pedal in the only spot where they could physically fit, then I assume I too am a genius for correctly inserting my legs into my pants this morning." - candletears7 - TGP
- IvIark
- Tube Twister
Information
This may help: http://www.electrosmash.com/jrc4558-analysis
"If anyone is a 'genius' for putting jacks in such a pedal in the only spot where they could physically fit, then I assume I too am a genius for correctly inserting my legs into my pants this morning." - candletears7 - TGP
- Nocentelli
- Tube Twister
Information
- Posts: 2222
- Joined: 09 Apr 2009, 07:06
- Location: Leeds, UK
- Has thanked: 1155 times
- Been thanked: 954 times
i think IvIark's right about doing an exact 4558D replica using the exact circuit with discrete components might be quite a task. i'm sure plenty of people have tried a discrete TS before eg
https://www.diystompboxes.com/smfforum/ ... 19.20;wap2
but that thread seems to suggest it was tricky to replicate the gain level.
https://www.diystompboxes.com/smfforum/ ... 19.20;wap2
but that thread seems to suggest it was tricky to replicate the gain level.
modman wrote: ↑ Let's hope it's not a hit, because soldering up the same pedal everyday, is a sad life. It's that same ole devilish double bind again...
- J0K3RX
- Degoop Doctor
- Attachments
-
- EE-MinimaxPlus-discreteOP-4.jpg (44.33 KiB) Viewed 7172 times
- IvIark
- Tube Twister
Information
That was it!!
"If anyone is a 'genius' for putting jacks in such a pedal in the only spot where they could physically fit, then I assume I too am a genius for correctly inserting my legs into my pants this morning." - candletears7 - TGP
- commathe
- Breadboard Brother
I find it tough to believe a discrete component op-amp will be better. You are adding in so many levels of inaccuracy and component variation that I am highly skeptical it will *imporve* performance.
That Burson discrete op-amps page made me facepalm so hard for so many reasons. It was the "let it burn in for 100 then enjoy" part that made me stop taking them seriously.
That Burson discrete op-amps page made me facepalm so hard for so many reasons. It was the "let it burn in for 100 then enjoy" part that made me stop taking them seriously.
- Blitz Krieg
- Breadboard Brother
i think this is where most people get hung up. you won't see many values because so much depends on what's going on around it. these are "textbook" designs, but the authors don't know what your current demands are, or impedance etc.differential amplifier with a current mirror
- FiveseveN
- Cap Cooler
Information
First of all, you need to distance yourself from that person. It's a stupid statement on so many levels!Mrdorbe wrote:someone planted a bug in my head saying that guitar pedals sounds amazingly better with no opamps or chips in them
If you plan on building a discrete op amp for educational purposes, go right ahead. Might want to start with some simpler design, though.
Ignorance more frequently begets confidence than does knowledge. (Charles Darwin)
- mirosol
- Resistor Ronker
A single opamp would probably be more viable. But. If you want to take it up a notch from this idea, you could also discard the idea of using transistors and swap those for diodes/caps/resistors.
I think the bug in your head originally generalizes the tought. That bug may have meant that discrete JFET circuits tend to sound "better" than opamp based ones. I bet you won't get Wampler's Plexi-Drive tones with any opamp. Still i personally think that recreating a 4558 with discrete components is rather silly idea :)
I think the bug in your head originally generalizes the tought. That bug may have meant that discrete JFET circuits tend to sound "better" than opamp based ones. I bet you won't get Wampler's Plexi-Drive tones with any opamp. Still i personally think that recreating a 4558 with discrete components is rather silly idea :)
http://tagboardeffects.blogspot.com/
http://mirosol.kapsi.fi/
"No such thing as innocence" -Iron Chic
http://mirosol.kapsi.fi/
"No such thing as innocence" -Iron Chic
- caspercody
- Resistor Ronker
- jymaze
- Resistor Ronker
A discrete opamp has not much reason to sound better than an IC. If you don't like the sound of distortion of high gain / high feedback devices, then it will sound the same to you when distorting. You should try cascaded jfets designs if you haven't tried yet (low gain / no feedback).
If I were you, I'd build a very simple disrete opamp for audio applications.
I made two of them, including the diode opamp (https://www.diystompboxes.com/analogalc ... opamp.html)which sounds quite good for distortion applications.
I made two of them, including the diode opamp (https://www.diystompboxes.com/analogalc ... opamp.html)which sounds quite good for distortion applications.
Patafix... so sticky
I agree with Patafix. There is no reason not to play around with your own discrete Op Amps. They are simple in concept, but they get really complicated when you add all the current sources and current mirrors. Leave all that out and you may create one that sounds as bad as a 4558!
I would start here:
https://www.passdiy.com/project/amplifiers/diy-op-amps
Download the pdf.
I would start here:
https://www.passdiy.com/project/amplifiers/diy-op-amps
Download the pdf.
I think it's API who use discrete opamps in their kit. Have a look at them maybe (there are loads of schematics around)
- Ice-9
- Degoop Doctor
Information
Have a look at the schematic for the Boss BD2 pedal, it is basically using the same circuitry that an op amp is comprised of but using discrete components. I personally doubt it is worth the effort to make a discrete version of an op amp other than 'snake oil' marketing bollocks.
Here is a discrete op amp product that will also plug directly into an 8 pin DIL socket, http://www.profusionplc.com/parts/V5-OPA-D at £50 each.
Here is a discrete op amp product that will also plug directly into an 8 pin DIL socket, http://www.profusionplc.com/parts/V5-OPA-D at £50 each.
It's fairly straight forward, if you want to start it , press start. You can work out the rest of the controls for yourself !
No silicon heaven ? preposterous ! Where would all the calculators go ?
No silicon heaven ? preposterous ! Where would all the calculators go ?
- mictester
- Old Solderhand
Information
Looking at that "Burson" discrete op-amp makes me certain that there's no fool like an Audiophool!
The transistors on a single piece of silicon are going to be extremely well matched - I would go so far as to say that they would be identical. This will also apply to the resistors in the circuit - they will be extremely precisely matched. If the components in the "Burson" effort aren't matched to within a couple of parts-per-million, the performance of their thing will be inferior to the 4558s I buy at 20 for a $1 in the Far East! Call me cynical, but I'd rather pay $0.05 for a dual op-amp than $80 for one - especially when the cheaper version will perform significantly better than the discrete copy!
The transistors on a single piece of silicon are going to be extremely well matched - I would go so far as to say that they would be identical. This will also apply to the resistors in the circuit - they will be extremely precisely matched. If the components in the "Burson" effort aren't matched to within a couple of parts-per-million, the performance of their thing will be inferior to the 4558s I buy at 20 for a $1 in the Far East! Call me cynical, but I'd rather pay $0.05 for a dual op-amp than $80 for one - especially when the cheaper version will perform significantly better than the discrete copy!
"Why is it humming?" "Because it doesn't know the words!"
- Ice-9
- Degoop Doctor
Information
mictester wrote:Looking at that "Burson" discrete op-amp makes me certain that there's no fool like an Audiophool!
The transistors on a single piece of silicon are going to be extremely well matched - I would go so far as to say that they would be identical. This will also apply to the resistors in the circuit - they will be extremely precisely matched. If the components in the "Burson" effort aren't matched to within a couple of parts-per-million, the performance of their thing will be inferior to the 4558s I buy at 20 for a $1 in the Far East! Call me cynical, but I'd rather pay $0.05 for a dual op-amp than $80 for one - especially when the cheaper version will perform significantly better than the discrete copy!
All the TI op-amps already use laser trimmed resistors and matched FETs (where used) and I expect other manufacturers op amps do as well . Discrete op amps are definite Snake oil for the HiFi brigade.
It's fairly straight forward, if you want to start it , press start. You can work out the rest of the controls for yourself !
No silicon heaven ? preposterous ! Where would all the calculators go ?
No silicon heaven ? preposterous ! Where would all the calculators go ?
Yes, API discrete opamps are pretty popular in the DIY "pro-audio" community. There are also lots of others around which are simple compared to a 4558 (and probably better documented). Lots of useful stuff here, for example:
https://bgilbertsound.com/2010/03/28/di ... onal-amps/
https://bgilbertsound.com/2010/03/28/di ... onal-amps/