Hello
I have a synthesizer which I enjoy playing, but it is lacking enough output volume compared to others. I am willing to build a small, simple stereo preamp for it, and I am looking for advices.

My needs would be :
- +12 to +18dB of clean boost, with no bass or treble attenuation;
- stereo operation;
- The boost device would not add (too much) additional noise;
- Maybe +9 to +18V for increased headroom (not mandatory);
- Maybe a "dirt" switch for some slight coloration when needed (not mandatory);

I am hesitating between transistor and opamp design, and I think there are plenty schmeatics and veroboard layouts out there, but I don't know which one to chose...

- I have considered building two EQD Black Eye (or two modded Red Llama) into a single stereo enclosure;
- I am searching the web for veroboards of stereo preamps, but I am mainly finding "audiophile" preamps so far...
- I have the feeling that I could mod this Klon-style buffer into a stereo booster, but I don't know how to do (and if it would be a good idea)

Would you have any advice or recommendation?
Thank you in advance!

My go to is a pair of MXR microamps, which can run on 18v.
http://www.tonepad.com/project.asp?id=28

hi, Idy
Thank you for your suggestion : This is also what occured to me after  browsing the web for a while 

I will investigate.
Maybe building a stereo version with a dual opamp ?
And maybe featuring a switch adding leds to the circuit, for a transparent/overdriven switch?

+1 what idy said. A simple dual op-amp (even on 9V) will give you minimal noise and can definitely achieve the clean boost you need.

Thank you, Electric Druid.
So let's go for a dual MXR MicroAmp then.
I am considering adding either diodes/leds on a switch, or an EQD Speaker Cranker on top of it when a slight dose of saturation is needed

I will suggest that you aim for optimum headroom by using a +/-12V supply.  Even if this one is low, by comparison to others, synths tend to put out some pretty hefty signals, applying 12db boost may exceed what a 12V supply can provide in the way of headroom.  And should you ever venture into the realm of Eurorack, you'll have a +/-12V supply ready.

It's a synth. Therefore, a low impedance output (I'm guessing, do you have any specs?). How low is the output? If you have a sine oscillator on it (I'm guessing a triangle would work fine too) then you can measure its output with a simple multimeter on AC mode (it will measure the RMS value, you need to multiply it by √2 to get the peak value, twice this for peak to peak - if the sine is not a sine, true peak values will be lower). +12 to +18dB means a gain of 4-8.

This means maximum output needs to be, at 9V, about 6V swing / 4-8. (Opamps clip usually at about the supply rail +/-1.5V, so subtract 3V from the supply voltage). That is, 0.75 to 1.5Vpp (about 0.26 to 0.52 Vrms).

18V supply instead: 15V swing / 4-8. Therefore 1.875-3.75Vpp (so, about 0.66 to 1.32Vrms).

18V makes it easier to use many opamps (many of which are not specified for operation at 9V) but unless you are looking for extremely low noise, and assuming a low source impedance, a bipolar input opamp such as the NJM4580 can be a good choice. It's cheap and it works at low(er) voltages. Otherwise the NE5532 is cheap and almost top of the class.

In both cases, assuming a low source impedance, the parts count is, just as a quick calculation... LED, 3.6k or so current limiting resistor, diode for reverse polarity protection, the opamp, the 0.1u MLCC bypass capacitor, two 220u caps, four diodes for input overvoltage protection (or not  ), two 10u input caps, four 10k resistors for biasing (two for the voltage divider to set Vref, two as biasing resistors - 10kohm input impedance per channel), then, another two resistors + a cap per channel for the feedback path (for instance, 2.7kohm + 47uF for the ground path and either a 8.2kohm or a 18kohm feedback resistor from output to Input-), a 100ohm or so resistor on the output of each channel, another two 10u caps on the output, two 100kohm resistors to ground after the caps. Four TS jacks (or two TRS ones) and a DC barrel jack. That's it. Should work fine. You probably should add in a 100p or so ultrasonic limiting cap in the feedback, parallel to the feedback resistor, per channel. All in all, about 30 components. Should expect about 0.01% THD or less. If you want complete flatness in band go bigger with the big caps (uF) and smaller with the small ones (pF), these are already overkill anyway. Oh, all caps should be rated for at least three times the DC voltage they will be subject to. If you're supplying 18V, then the 220u power supply cap should be at least 50V, all the others 35V. (25V can work fine too).

Don't know if you can make any sense of what I said. Also, I might be talking bollocks. I'm an amateur.

PS: if you go with a dual supply - but you'll probably have to build it yourself, and you might as well go with +/-15V - there is no Vref to set but another 0.1uF MLCC bypass capacitor from V- to ground. If you're building your own power supply, you might want to use voltage regulators.

While I understand people's desire for more headroom, the initial problem was that the synth is quiet. If it were boosted to the point just before it starts to clip on a 9V supply, we'd be looking at about 6Vpp output, or perhaps a touch more. There's *no way* a 6Vpp signal is "quiet" in any normal situation, so moving to 18V headroom seems like overkill. Even +4dBu "Pro" line level is only 3.5Vpp, so 6Vpp is more than enough to whack the inputs of most gear.

About the only exception I can think of is if you're feeding it into a modular synth that's expecting +/-10V (20Vpp) levels. In that situation, it might be a bit quiet!

I suppose the other question is whether there might be something wrong with the synth that's making it quiet? Dodgy volume control? Bad output? It could be we need to *fix* it, rather than boost it.

FWIW, powering a NE5532 (and alike) with 36V (+/- 18V) supply is like you're asking for trouble..

Thank you all for your replies.
I will try to answer everything :

@Mark : Yes, I do have a small Eurorack system with a symmetrical +/-12V power supply. However, in the studio the synth will not be very close to the Eurorack (and I wouldn't want to use the whole modular system just for beefing up the Blofeld weak output level). I'd rather build an internal +/-12V daughterboard if needed.

@Fryingpan : Wow, thank you for all the technical information!

Yes, NE5532 is what I had in mind (this, or either a TL071). I have several 5532 in a bag, so I think it is what I am going to use.

About the capacitors : according to this website, there should be no need to increase the size of input capacitors



@ElectricDruid : Yes, that's it. The synth is too quiet. It's rather boring, because it's really not on par with the other synths in the studio, so I'm willing to build a small "set and forget" boosting device so I won't have to worry about it anymore.
It's a Waldorf Blofeld. It's a lovely synth, but the weak output level is a known issue (or "feature?"), and I don't think there is anything which can be done to improve it.

Quote from: snk on October 01, 2024, 02:55:59 AMThank you all for your replies.
I will try to answer everything :

@Mark : Yes, I do have a small Eurorack system with a symmetrical +/-12V power supply. However, in the studio the synth will not be very close to the Eurorack (and I wouldn't want to use the whole modular system just for beefing up the Blofeld weak output level). I'd rather build an internal +/-12V daughterboard if needed.

@Fryingpan : Wow, thank you for all the technical information!

Yes, NE5532 is what I had in mind (this, or either a TL071). I have several 5532 in a bag, so I think it is what I am going to use.

About the capacitors : according to this website, there should be no need to increase the size of input capacitors



@ElectricDruid : Yes, that's it. The synth is too quiet. It's rather boring, because it's really not on par with the other synths in the studio, so I'm willing to build a small "set and forget" boosting device so I won't have to worry about it anymore.
It's a Waldorf Blofeld. It's a lovely synth, but the weak output level is a known issue (or "feature?"), and I don't think there is anything which can be done to improve it.

I have that very synth. It's quiet. Also a bit noisy. A great synth otherwise.

If you use 5532s you can get away with 12V (single supply). Of course not 18V dual, the most you can do with an NE5532 is 15V dual (30V single).

If you want some dirt, you could add two sets of two 1N4148 series / antiparallel diodes, maybe with a 22n cap before or after the antiparallel pair, and a switch. Of course, in the feedback path of each opamp. This way, you'd be looking at some mild soft clipping over 400Hz or so. (If you go with +18dB gain, if you go with +12dB a 6.8n cap works).

Quote from: fryingpan on October 01, 2024, 04:35:44 AMOf course, in the feedback path of each opamp.

Not necessarily..

Of course, it's a matter of taste but, IMHO, that set of diodes with a series resistor (for off-seting clipping threshold) from op-amp out to GND could "sound" better in such a preamp configuration..
(at least, for guitar/bass..)

Quote from: antonis on October 01, 2024, 05:24:45 AM

Quote from: fryingpan on October 01, 2024, 04:35:44 AMOf course, in the feedback path of each opamp.

Not necessarily..

Of course, it's a matter of taste but, IMHO, that set of diodes with a series resistor (for off-seting clipping threshold) from op-amp out to GND could "sound" better in such a preamp configuration..
(at least, for guitar/bass..)

Yes, I employ it myself in a number of designs... The series resistor could be anything from 2k to 10k (or beyond).

Ok, I will be building the circuit this evening or tomorrow.
Once built, I will consider adding a switch with diodes, but since I already have several dirt and coloration devices, my main goal here is to make the synth "just louder", so I might decide to keep it simple

One question : the Electrosmash website states that

QuoteThe 500K potentiometer R5 is Reverse Log. You can still use a Linear potentiometer but the action of it will not be evenly spread across the whole pot range. As you can see in the table below, everything occurs between 0 and 50K (and thats why a Reverse Log pot is needed):

 R5(kΩ)   GAIN
0            21.7
50          2.06
100        1.54
150        1.36
200        1.27
250        1.22
300        1.18
350        1.15
400        1.13
450        1.12
500        1.11

So, wouldn't it be best to use a 100K pot ?
https://www.electrosmash.com/mxr-microamp#link12

I am struggling to find a dual gang antilog 500k pot, and since my synth already features a volume pot, I think it might be easier and smarter to go for a fixed boost (or two values on a switch).

Am I right assuming that replacing the 500k antilog pot by a 12K resistor would give me roughly 12dB of amplification, and that a 5.1K resistor would give me +18dB (gain x8)?

with Gain = 1 + 56k/(R5+2k7)

Quote from: snk on October 01, 2024, 01:06:05 PMAm I right assuming that replacing the 500k antilog pot by a 12K resistor would give me roughly 12dB of amplification, and that a 5.1K resistor would give me +18dB (gain x8)?

with Gain = 1 + 56k/(R5+2k7)

Yes, that's right.

Here's an alternative approach using linear pot. A dual 100K linear pot should be easy to find.





Ok, it's not totally even, but it'll be fine for +12 or +18dB. It gives you a bit more control for the lower gain settings.

Not again.. 
(lost post..)!!!

Gagh, PostImg is not playing ball at all...

Let's try again...





Aha! That's better!

Oh, now *everything* suddenly appears...

Oh well.

Quote from: ElectricDruid on October 01, 2024, 03:48:19 PMOh, now *everything* suddenly appears...

Yours *everything*, not mine..!