Microphone pre amp for PTT

[nhoj_yelbom]

this is for peltor comtac III's and TCI TABCIII radio ptt headsets. the mic ohms out at 150.

i have made this circuit for the mic to work with my radio
http://www.epanorama.net/circuits/dynamic_to_electretinput.html

i get horrible motorboating noise (digital only, analog fine)
at first i thought it could be the cheap kenwood k1 cable i harvested off a baofeng ptt, but i got a new kenwood oem ptt cable on it now thats shielded

only thing it can be now is the pre amp im using to make the mic work, any suggestions or better circuits to try?
   

[nhoj_yelbom]

Bump

[orolo]

I think the problem with digital signals might be that this amplifier has disproportionate gain at higher frequencies, if I'm not mistaken. If the collector is biased by 5V via 2.2k, typical for a mic, the base current is about 6uA, and that means that the input impedance of the transistor is over 4K. That, together with the 220n of the cap produces a high pass filter at the input with about 1KHz 3dB point. If your digital data rate is well below 1KHz, you will hear only the step transitions of the square waves. A sinewave of about 100Hz should be greatly attenuated compared with a 2KHz one; that's a good check to know if this is the problem.

If the problem is the high pass filtering, you can try drastically increasing the 220n cap (say, 47u), together with a resistance in series of about 3-4K. That should take the 3db point quite close to 1Hz, without impairing the gain too much. Or you could try a different amplifier.

[nhoj_yelbom]

thank you. im open to using a different amplifier, just dont know what would be simple and good for this application. i dont have a scope to check the sinewaves, just increasing the capacitor size and put the resistors where?

[orolo]

If I'm right, the difference should be audible. Can you feed the audio, say from audacity with your computer, into the mic, and hear or record the output? The resistor should be in series with the cap, just at the base of the transistor. This is just a hack; I think this preamplifier is a rather crude design. I think a very high input impedance preamplifer, like a FET, would work, if you are trying to convert a low impedance, low voltage signal into a higher voltage, high impedance signal.

[nhoj_yelbom]

the audio goes through, but if you did not know the song playing it would not sound like music. i noticed the motorboating is not in the audio. i have not had a chance to hook the output up to record yet

[PeterFW]

When it comes to audio stuff, my first stop is allways Elliots website.

http://sound.westhost.com/

The next best thing i found that might do what you want is this one i think:

http://sound.westhost.com/project13.htm

[nhoj_yelbom]

Thank you, but dont think anything i see there can help for my application

[orolo]

Hi. Have you tried increasing the capacitance? The added series resistance should also cutoff the amplification over about 100KHz, in case your cable is picking up interference, even if your amp losses amplification: it is optional, and if you are not that worried about pickup, could be removed / reduced (try with a pot).

Project 13 as linked by PeterFW is a beautiful preamp, similar to yours (note R=1.2K and C=10u at the input of the first transistor, though in yours I'd place the resistor just at the base, for better gain), but with very clever multiple feedback going on. If you can adapt this circuit to your 5V (I guess) supply, it will beat the open loop design in all respects.

I'm not sure what do you mean by digital signals: is it Morse code, low baudrate digital audio, or what.

Just to clarify what do I mean by high pass filtering, I will attach a spice simulation of the preamp. It confirms most of the calculations I did by hand, with a little variation in bias (I assumed beta=100, that's low for a 547B).

The file preamp.png is the circuit as described. In gain.png you can see its frequency response: the gain rises at 6dB per octave below 1KHz; low frequency input is very much attenuated compared to higher frequencies. In squre_resp.png you have the output for a 10Hz square wave in this amplifier: you only hear the steps as a series of clicks. Could that be the motorboarding?

The file preamp_new is the improved version. In gain_new.png you have its response: you lose 6dB due to the resistor, but gain starts to decline at about 100KHz, if RF pickup is a problem. Remove the resistor if it isn't, you'll keep the gain, up to HF. Square_new.png is the amplification of the 10Hz square wave, far from optimal, but at least adequate.

I hope this helps.

[nhoj_yelbom]

i will have to get some parts to test those circuits. when i mean digital im talking about the radio using DMR @ 400-480MHz frequencies UHF. when i operate the radio on analog there is no issue. as soon as i get the items i will report back with the results! thank you

what are the V1 and V2 items?

[PeterFW]

what are the V1 and V2 items?

Those are the LTSpice voltage source symbols, that is the "power supply".

[nhoj_yelbom]

What about the 43u cap? When i search for one it comes up as a giant one? Sorry for the dumb questions....

[PeterFW]

What about the 43u cap? When i search for one it comes up as a giant one? Sorry for the dumb questions....

That is a 43µF capacitor, the next best thing would be two 22µF in parallel to get that value.

[nhoj_yelbom]

all i could find local is a 47uf and 4.7k resistor instead of 43uf and 4k

[orolo]

all i could find local is a 47uf and 4.7k resistor instead of 43uf and 4k

Sorry, it was a mistake. 47u should work as well. For the resistor, the best would be a 5K or 10K pot to trim the gain, else I would settle for the first practical value below 4K, like 3.9K, 3K or 2.2K. 4.7k should work as well.

[Audioguru]

The 43uF or 47uF input capacitor value is way too high. The mic is used for voice, not for recording earthquakes. The low frequencies response cuts off below 1Hz. Try 1uF or 0.47uF.

[orolo]

Yeah, when I thought digital, I thought very low frequency square waves, like those you might use in CW. That's why I asked about the frequency of the digital signals. I wasn't sure if the OP was using a mic to transmit higher frequency digital data via MIC. By the way, if I understood it right, this specific circuit will reach -3db at 150Hz, and with 1u at about 50Hz, with very unequal gain below. I'd definitely go for 10u at least. Thinking back, 1Hz cutoff is exagerated, but I don't think 10 is.