Mic Preamp -- Is This a Silly Idea?

[bcarso]

The case insensitivity of circuitmaker spice has tripped me up many a time. It's especially pernicious since I use it as a drafting program and in final document stages try to revert to capital M for meg. Often one or two on a large schematic are missed, which then screws up the extracted BOM.

 

[JohnRoberts]

I really hate it when computers try to help us, when the rules don't apply to the task at hand.

I sometimes run into problems with a graphics design program that is compelled to capitalize the first letter of any string of words...

Computers truly are dumb as silicon....

JR

 

[PRR]

> Most SPICEs are case insensitive

They go back to punch-cards, and low-price keypunches (the kind they let EEs use) might not have lower-case.

But no problem. There were no milli-ohm parts back in the stone age. If you had a milli parasitic, you wrote it in E notation, 22E-3, or just 0.022

And then some fool allowed millis into SPICE. And took advantage of these newfangled terminals with both upper and lower case.

So while most bits are case insensitive, "m" and "M" are a billion to one apart.

There is another SPICE-ish library which does not accept "K". Idiot only takes "k".

 

[clintrubber]

[quote author="PRR"]There is another SPICE-ish library which does not accept "K". Idiot only takes "k".[/quote]
Imho that last one actually doesn't sound too bad to me, kind of teaches people to use lowercase 'k' outside sim-city.

... unless they want to express values of components in two units at once... R_101 = 2.7 KelvinOhm ?

Regards,

Peter

 

[Harpo]

[quote author="chris319"]According to the TINA-TI simulator

this circuit turns 300 uV AC into 41.54 uV AC with the addition of C2 and R3. Without C2 and R3

Fixed that, thanks. Now 300uV AC becomes 6.61 V AC (+86.9 dB). Note the use of the much less expensive OPA134.

[/quote]
what is the use of C1 being 7,5pF ? A lowpassfilter at 2,1MHz seems a little on the high side. I'd expected something like 82-100pF in this spot.
I'd also divide the value of C2 (or R3) by 10, causing phase error >-5° only below 18Hz. YMMV. Maybe I am missing something.
-Harpo

 

[chris319]

Do I even need C1? It was part of a TI circuit design.

 

[Harpo]

[quote author="chris319"]Do I even need C1? It was part of a TI circuit design.[/quote]
Yes. IIRC Samuel already explained it to you in a previous post. As you seem to be looking for spots to save some money, I mentioned the possible lower value of C1. (at some expense you can build a servo circuit to catch the expected offset in order to get rid of C1 in this spot).
From your schematic the 2nd gain stage will not add your quoted 20dB. Recheck your math for log10(1+R2/R1)*20.
I still don't see much sense in a 86dB gain stage (actually you have even more on paper) for field recording.

 

[jdbakker]

[quote author="Harpo"][quote author="chris319"]Do I even need C1? It was part of a TI circuit design.[/quote]
Yes. IIRC Samuel already explained it to you in a previous post. As you seem to be looking for spots to save some money, I mentioned the possible lower value of C1. (at some expense you can build a servo circuit to catch the expected offset in order to get rid of C1 in this spot).[/quote]
Harpo, I believe that's C2 you're talking about. C1 rolls off the gain of the second stage at higher frequencies, in an attempt to keep your mic pre from turning into (a) an AM band receiver or (b) an AM band transmitter. Chris, you scaled R2 down by a factor of 30; to remain effective, you'd want to scale C1 up by a similar factor (ie to ~220pF). That way your second stage has a 6dB/octave rolloff at supersonic frequencies with a 3dB-point of ~70kHz.

JDB.
[it's not entirely trivial to keep an amp with 80+dB amplification from turning into a radio transmitter/oscillator]

 

[chris319]

Harpo is right; R1 needs to be 750 ohms for 20 dB of gain. C1 is now 220 pF. There is also a 47-ohm series resistor on the output. There will also be bypass caps which are not shown in this drawing. What do we think now?

I still don't see much sense in a 86dB gain stage (actually you have even more on paper) for field recording.

I'm old enough to remember when dialog was recorded and broadcast using RCA 77DXs, BK5Bs and E-V 642s and RE15s hung on mic booms. The added gain makes all the difference, IMO. With each 6 dB of gain you can work the mike at double the distance. With 18 dB 3" become 24".

keep your mic pre from turning into (a) an AM band receiver

This is not lost on me. There is a 50,000 watt, six-tower directional antenna at 1580 kHz just miles from here and the beam is aimed straight at me! The issue of ingress should not be taken lightly.

 

[Samuel Groner]

R1 needs to be 750 ohms for 20 dB of gain.

Err, 750 ohm gives ~23 dB, 1k1 is much closer to 20 dB...

I would set C1 to 100 pF, 220 pF gives (as mentioned) a 70 kHz roll-off which seems a bit low for my taste.

Otherwise it looks fine to me.

Samuel

 

[chris319]

I finally found the time to build my two-stage preamp:

Instead of an INA163 there is a THAT 1510 with phantom power protection diodes and resistors ahead of the input on the chip. They are not shown in the schematic.

Tested individually, each stage gives satisfactory THD, but combining stages results in a sizable increase in THD according to RMAA (I will post specific numbers later) but going from memory it goes from, say, 0.00xx% to 0.0xxx%. For C2 I am using a ceramic capacitor which I happened to have on hand (need to order some polypropylenes). Could this account for the rise in THD?

The THAT 1510 performs very well and delivers up to 66 dB of gain by actual test. The second stage is set for 20 dB of gain. The noise floor goes up when the stages are combined but that is to be expected. The frequency response is very good both for each individual stage and for the combined stages.

Many thanks.