Maybe because all the clipping/distortion comes from FET before output pair?
Getting capsule polarization voltage from phantom power in a Schoeps-style circuit
- Thread starter skyy7
- Start date
Help Support GroupDIY Audio Forum:
thor.zmt
Well-known member
No, it's the P12 12V Phantom power thing.
If you take the design and optimise for 48V (without consideration of polarisation voltage by using a classic step-up oscillator for that) you get reduced distortion at higher (non-clipping) levels vs the original and an onset of clipping at much higher levels.
That was the first step that led in steps to my all FET design with capsule bias from phantom power and P68 option with 60V capsule bias.
Incidentally, we have tiny modern switchers that operate at 1Mhz+ that can be made to work for the capsule bias. I'm surprised nobody uses them. 1MHz is very easy to filter out and the whole circuit using SMD parts is tiny. Only reason I didn't do this was the unwillingness to design and spin a PCB for a few mikes when instead a few minor mods on the desk give me P68.
Thor
Matador
Well-known member
I have only designed circuits with MOSFETs as switches; my first thought when looking at that FET variant of the Schoeps style circuit is that the variability in FET threshold voltage would make DC coupled circuits like that difficult to match side to side. How does that work out in practice?
thor.zmt
Well-known member
Modern parts have very good consistency. Mosfet's and J-Fet's. Parts from the same reel commonly are essentially identical. Panta Rhei, Tempus Fugit.
I never observed material offset, given the inputs are usually AC coupled anyway it matters zip.
Thor
thor.zmt
Well-known member
I was more thinking something like this:
Axelite AX5511
This tops out at ~ 24V so we drive a greinacher cascade, capsule bias draws zero actual current once stable, so 1N4148 and 100n X7R ceramic cap's will be plenty. With regulation set to 66V and a tripler cascade we get 66V Bias with 22V at the switching node, all well within safe parameters.
The rest is still about an optimised Schoeps topology.
We only need 5V input for the switcher (note, there is a trick to use more than 5.5V input, drive the inductor from a higher voltage, the chip's Vin pin from a divider with zener clamp), so we can get (say) 32V across our 6.8k Resistors (9.5mA total current), 11V across the output followers for around +26dBu at hard clipping, giving some headroom.
We use the main 22V from the switcher to drive our J-Fet frontend at the usual 0.5mA or so (we have up to ~1mA available.
This incidentally places the J-Fet gate at ~ 5.5V, hence the 66V bias to get actually 60V bias (66V - 5.5V). The input resistance/diode is bootstrapped in this case, no gate resistor, capsule straight to gate.
Our frontend can handle around +20dBu output (bal) before hard clipping. With ~ -32dBu/Pa (20mV @ 94dB) we can theoretically handle 146dB SPL at the point where electronics clip.
Thor
Last edited:
Matador
Well-known member
I had breadboarded up a test circuit, using a OPA210 (lower noise than your choice, although at higher current), and without phantom power (just using a test supply that provided +- 15V, and also +60V for the capsule). The chip was in SOT23-5 which allowed the non-inverting pin to be bent upward into the air for connecting to the 1G / capsule point.
The main different to your implementation is that it used a second op-amp to add a LPF to tailor the HF response. If I recall I preferred a 7kHz pole, however that was using a K67 copy that was quite shrill at HF.
But other than that, I thought it sounded great.
rogs
Well-known member
I hadn't realised that the OPA210 had an input impedance high enough for this task?... Although the OPA164* series is a little noisier, it does have a JFET input, making it ideal for condenser capsule interfacing.
I've not found the noise floor a problem -- the ambient noise tends to swamp the op amp noise in most 'real world' applications.
I know that some folk like to experiment with EQ within the mic, but I have found that adding resistors into passive filters in the signal path at mic level can tend to get a little bit noisy..... But if you're trying to copy a specific mic schematic, for example, it is of course entirely appropriate.
I prefer to try and just use a single op-amp unity gain buffer as an impedance converter, with no extra resistors in the signal path, and then add eq etc outside the mic.
At the moment I'm doing a dual polarity DC voltage multiplier, so I can get rid of the sum and difference amps in my multi-pattern OPIC mic.
Even simple functions like adding and subtracting place extra resistors into signal path -- and they add noise (Not a lot -- but it's still extra noise!)
I'm currently experimenting with a Dickson charge pump ... Nice simple reliable concept that should allow a simple multiplier with an adjustable range of ± 60 to 80V.... And no 'inductor coupling' orientation problems to worry about!
Matador
Well-known member
You are correct! The part number was OPA1656, I just looked at the wrong part in my stash. It looks largely the same as what you tested. I think it even worked with a TL072.
I'm also interested in placing the SOT-23 op amp and 1G resistor on a tiny PCB located up in the head basket to completely remove the high impedance parts of the amp from the shell to the head basket.
TLC072 is used in Townsend Labs Sphere. Note it's TLC not TL.You are correct! The part number was OPA1656, I just looked at the wrong part in my stash. It looks largely the same as what you tested. I think it even worked with a TL072.
I'm also interested in placing the SOT-23 op amp and 1G resistor on a tiny PCB located up in the head basket to completely remove the high impedance parts of the amp from the shell to the head basket.
Very different animals in this application. TLC has 15dB less noise current, 8dB less noise voltage.
