Mesa GEQ Inductors - Gyrators

[mensur]

Hi guys,
I'll try to replace mesa inductors with gyrators, but I need some help:
1.Is the transistor emitter bias resistor fine(6.8K), or I should use another value?
2.Do I need to use dual voltage supply for the circuit?



Thanks

[loudthud]

I'm not sure you have the right form of gyrator. Doesn't the Mesa circuit have slide pots with series LC tanks to ground? Why does your circuit have In's and Out's? The circuits I've seen simulated the inductor to ground and the cap blocked DC so a split supply wasn't required. You just need to make sure the gyrators don't run out of current and that depends on how much signal the equalizer needs to handle.

[mensur]

Thank you, here is the complete schematic:

You telling me to use single voltage supply if I put inductors before the caps,or the caps will block DC anyway, so I can put gyrators directly to ground?
And what about Q-point resistors (100R's)?

[teemuk]

The nodes you have labelled as "out" are not outputs. They should connect to reference voltage, equal to approximately half of the total power supply voltage. In dual supply circuits you can simply connect them to ground but if you use a single-rail supply and ground the emitters the nodes must connect to Vcc/2, most conveniently made by a resistive divider with the bottom resistor AC bypassed. Fortunately all gyrators can likely use the same DC reference to simplify the circuit.

As far as I know, these kinds of gyrator circuits need to have one end grounded. Since the circuit in the EQ is a series resonance circuit it's easy to arrange simply by changing the order of the components in the series string. For example, place the capacitors (connecting to ground in your schematic) before the gyrator inductors. The gyrators will ground the strings.

[Steve Conner]

Originally Posted by teemuk

As far as I know, these kinds of gyrator circuits need to have one end grounded.

Yes, they do. There's another gyrator circuit that allows both ends of the "inductor" to be floating, but it's much more complicated. It's not needed in this application anyway.

Look at RG Keen's EQ article to see how it all wires up.

[mensur]

Thank you Teemu that's all I needed to know,
Here's V2 of the schematic:


Let me know please If i need to change something.

AND CAN USE 100R RESISTORS FOR THE Q-POINT?

[Steve Conner]

Teemu might know better, but I think the 100R resistor will be OK as long as it's large compared to the transistor's internal emitter resistance. That's (26/Ic) ohms, or whatever. If it's not, you either need a bigger resistor, or more collector current.

You can make the same circuit with an op-amp follower instead of the transistor, it's easier to design because the op-amp is closer to an ideal follower. It may even take up less board space if you use quad op-amps.

[mensur]

I know that its easier to design with op-amps but I don't like them, I want to make discrete gyrators.Can I put 1K's instead of 100R's and decrease 100nF's to the 10nF's, or out resistors (100K, 39K, 22K to 10K, 3.9K, 2.2K...)? The value of inductance will not change.

[Steve Conner]

The inductance won't change, but the Q will decrease because of the added series resistance. That's probably why that point is called the Q point. The result will be that you'll get less dBs of boost/cut than if you'd used 100R.

I find a simulator like LTSpice very useful for simulating EQ circuits like this.

[Gregg]

I'll try to replace mesa inductors with gyrators, but I need some help

I did this some time ago. Check out BA3812L's data sheet (or similar opamp EQs) and take the resistor values from there. Create a V/2 bias point for the base resistors and recalculate filter frequencies according to the formula.
I have a PCB image if you're interested. It was designed for 30mm travel slide pots to be soldered directly to the board but can be used with other pots as well.

[mensur]

I searched the schematics of different GEQ's and I found that typical value for Q-point resistor is 330R, 470R on Op-amp based gyrators and 1K transistor based gyrator on DOD FX-40B. I found 100R on Rod Elliot's example of gyrator,so maybe he was just using it for easier calculations as a example.

What will change If
1.R1-Q-point-1K, R2-10K, C1-100nF?
2.R1-Q-point-1K, R2-100K, C1-10nF?
Inductance and frequency wont, but what will happen with signal,does it mean that gain will be smaller with 100K,because of larger series resistance?

Thanks guys.

[Gregg]

I searched the schematics of different GEQ's and I found that typical value for Q-point resistor is 330R, 470R on Op-amp based gyrators and 1K transistor based gyrator on DOD FX-40B. I found 100R on Rod Elliot's example of gyrator,so maybe he was just using it for easier calculations as a example.

What will change If
1.R1-Q-point-1K, R2-10K, C1-100nF?
2.R1-Q-point-1K, R2-100K, C1-10nF?
Inductance and frequency wont, but what will happen with signal,does it mean that gain will be smaller with 100K,because of larger series resistance?

Thanks guys.

Maybe this will answer your questions:

Guitar Pedals: Gyrator Filter Calculator