Attached is a sample of a preamp/tone control circuit.
I the Tone control area is similar to the one I intend on using, however I will be using a Rod Elliot design instead.
The TL072 is the op amp I will be using.
B+ and B- will be 12/12v, regulated.
My question is...
The first op amp (IC1A) in the feedback line, R23/RF 1K..... if instead I replace R23 with a 1.5K in series with a 10K trimmer pot, will it give me a reasonable 'level' adjustment range, and how much?
I want to be able to use it as a 'preset-balance' control to match R/L channel levels in a stereo amp, and also compensate for weak line/source levels, maybe a 10dB variable range?
Also, I intend on feeding the pin3 of IC1A directly from a line level source, no volume pot there, just a 1uF DC blocking cap with a 100K to ground input.
The main volume pot will be located after the tone control section.
I the Tone control area is similar to the one I intend on using, however I will be using a Rod Elliot design instead.
The TL072 is the op amp I will be using.
B+ and B- will be 12/12v, regulated.
My question is...
The first op amp (IC1A) in the feedback line, R23/RF 1K..... if instead I replace R23 with a 1.5K in series with a 10K trimmer pot, will it give me a reasonable 'level' adjustment range, and how much?
I want to be able to use it as a 'preset-balance' control to match R/L channel levels in a stereo amp, and also compensate for weak line/source levels, maybe a 10dB variable range?
Also, I intend on feeding the pin3 of IC1A directly from a line level source, no volume pot there, just a 1uF DC blocking cap with a 100K to ground input.
The main volume pot will be located after the tone control section.
I think you'll see around 0.1db gain to just under 1db with a 10k pot and fixed 1.5k
Half of a quarter of nothing 😉
If you fix R23 at 10k and add a 100k preset in series with a 4k7 in the lower limb you can get 0.8 db to 10db range approx. Ideally AC couple the feedback return although with the TL072 (FET so essentially zero bias current) the preset should be noise free.
Half of a quarter of nothing 😉
If you fix R23 at 10k and add a 100k preset in series with a 4k7 in the lower limb you can get 0.8 db to 10db range approx. Ideally AC couple the feedback return although with the TL072 (FET so essentially zero bias current) the preset should be noise free.
Like this. The cap can be a 10uF if you want to include it. That would give a -3db point way below 10Hz but with a FET opamp you should really should not need it in practice.
That's it for tonight 🙂
That's it for tonight 🙂
Well, I changed stuff around and now have THIS modification on the board.
I hope that it gives me what I orginally posted here.
It'll be a while before I have it set up to test..
Others welcome to chime in.
I hope that it gives me what I orginally posted here.
It'll be a while before I have it set up to test..
Others welcome to chime in.
I'd put a resistor in series with the trimpot as in @Mooly's drawing, otherwise the op amp will run open loop if the trimmer is turned all the way down. 4.7k yields a 10dB adjustment range.
You also need a DC path to ground at the noninverting input. 100k resistor to ground after C18 should be fine.
You also need a DC path to ground at the noninverting input. 100k resistor to ground after C18 should be fine.
A slightly different circuit: make R23 the 10K pot, R24 = 4.7K, and connect wiper to opamp inverting input. Gain will vary between 1 and 3.128, i.e. 9.9 dB gain variation.
I would run the chip at +/- 9V, and put a 4k7 trim pot in the input section.
A 12V linear wall wart, with split ground in the circuit, gives about 17V in unloaded condition, this one will draw milliamps.
So +/- 8.5V, enough.
I found that you need to change input resistance to match phono input.
You might try a trim pot at the output side, leave it alone after matching, use a separate volume pot.
A 12V linear wall wart, with split ground in the circuit, gives about 17V in unloaded condition, this one will draw milliamps.
So +/- 8.5V, enough.
I found that you need to change input resistance to match phono input.
You might try a trim pot at the output side, leave it alone after matching, use a separate volume pot.
I appreciate the feedback and suggestions, however the board's already been populated, and any further changes will damage the foils and cause a mess.
The tone control circuitry is on the board and I don't want to go through making a whole new board.
The tone control circuitry is on the board and I don't want to go through making a whole new board.
WiseTech doesn't seem to need a phono input. You must be a vinyl fan? Phono inputs needed a lot more gain and RIAA EQ, and a 47K input for mm carts.
The gain of the input buffer is =(R23+R24)/R24. And the + input DC path goes to ground reference through the 100K (move R4) or RV1A pot, and the - input to the output through R23 so the the op-amp DC output copies the ground reference. If you can live with a few mV of offset, then OK, skip the capacitor. If this were not an FET op-amp, the resistance of R23 vs the RV1A pot/100K would matter, and depending on RV1A slider to remain connected is a bad idea and should be paralleled with a fixed resistor.
Any gain ahead of the volume control is in danger of being overdriven, so I don't think reducing the supply to 9V is a good idea. I would use an adjustable inverting amp so that no input could overload it and the gain can be adjusted say ~+/- 20dB. An inverting amp is also very robust with ESD and RFI, and they go quiet with high-Z input, and possibly better THD. Yes, the noise is a tiny bit worse but should not be a problem at line level.
The Baxandal EQ can be done with two caps instead of 4, but the results may not be exactly the same. You don't show the volume control which is normally after all at the output; the advantage being that noise comes down with the signal.
Today, balance controls are gone, and in many cases, like computer speakers, the only tone control is the level of the subwoofer. Most media players support an number of EQ settings.
The gain of the input buffer is =(R23+R24)/R24. And the + input DC path goes to ground reference through the 100K (move R4) or RV1A pot, and the - input to the output through R23 so the the op-amp DC output copies the ground reference. If you can live with a few mV of offset, then OK, skip the capacitor. If this were not an FET op-amp, the resistance of R23 vs the RV1A pot/100K would matter, and depending on RV1A slider to remain connected is a bad idea and should be paralleled with a fixed resistor.
Any gain ahead of the volume control is in danger of being overdriven, so I don't think reducing the supply to 9V is a good idea. I would use an adjustable inverting amp so that no input could overload it and the gain can be adjusted say ~+/- 20dB. An inverting amp is also very robust with ESD and RFI, and they go quiet with high-Z input, and possibly better THD. Yes, the noise is a tiny bit worse but should not be a problem at line level.
The Baxandal EQ can be done with two caps instead of 4, but the results may not be exactly the same. You don't show the volume control which is normally after all at the output; the advantage being that noise comes down with the signal.
Today, balance controls are gone, and in many cases, like computer speakers, the only tone control is the level of the subwoofer. Most media players support an number of EQ settings.
My experience comes from building a phono stage, as the intended amplifier did not have one.
The unit had to be small, and use easily available parts.
I used a TL072, and a wall wart, and had to change the input and output resistors.
The 072 can go much higher, but 18 Volts means custom made supply.
It worked, still does.
Options sold locally were 4558 and 5532, 072 had the fastest slew rate. So I used it.
I did not do any design, not worth the hassle for a one-off, just adapted the circuit from my Kenwood amp, that was for a 6556 or something chip, so I had to change resistors...that runs on 14V AC, you can work out the DC used.
The unit had to be small, and use easily available parts.
I used a TL072, and a wall wart, and had to change the input and output resistors.
The 072 can go much higher, but 18 Volts means custom made supply.
It worked, still does.
Options sold locally were 4558 and 5532, 072 had the fastest slew rate. So I used it.
I did not do any design, not worth the hassle for a one-off, just adapted the circuit from my Kenwood amp, that was for a 6556 or something chip, so I had to change resistors...that runs on 14V AC, you can work out the DC used.
This preamp/tone circuit will be fed by a 3 position selector switch. "phono-cd-aux".
Yes, there is an RIAA preamp feeding the selector, the cd-aux are direct to rca jacks.
After the tone IC, comes the volume, balance control, a preamp out option, then to the main amp.
Traditional, simple.
Yes, there is an RIAA preamp feeding the selector, the cd-aux are direct to rca jacks.
After the tone IC, comes the volume, balance control, a preamp out option, then to the main amp.
Traditional, simple.
Same as my Kenwood...it has an extra pre-amp for phono, the other inputsare direct to a single pre-amp, and its tone / balance circuits.
The phono signal is amplified by the first one to meet the needs of the second chip.
1988, which can be called traditional now...and of course simple, like many Japanese designs.
Ever see such stuff in Marantz, for example?
The phono signal is amplified by the first one to meet the needs of the second chip.
1988, which can be called traditional now...and of course simple, like many Japanese designs.
Ever see such stuff in Marantz, for example?
Bit more time now...
Here is the gain with the preset at 100k, 50k and 'short'. You can see at 50k the gain is only slightly higher than at 100k. Lower trace is just phase. You could also use a lower value for the preset such as 47k or even 22k. 22k gives a minimum gain of around 2.7db, same maximum gain of course.
The second image is the preset at 6k which gives a mid gain value. If it is just used for balancing then it should work well as you will get fine adjustment over quite a large rotational angle.
Here is the gain with the preset at 100k, 50k and 'short'. You can see at 50k the gain is only slightly higher than at 100k. Lower trace is just phase. You could also use a lower value for the preset such as 47k or even 22k. 22k gives a minimum gain of around 2.7db, same maximum gain of course.
The second image is the preset at 6k which gives a mid gain value. If it is just used for balancing then it should work well as you will get fine adjustment over quite a large rotational angle.