Need help troubleshooting a Pimped Alice circuit

[skyy7]

Hello, I am currently building my first microphone. It's a "Pimped Alice" by Jules that can be found here: Modify a Cheap LDC Condenser Microphone.

The circuit was made on perf board with a JLI-3412 capsule. I have made a few modifications by swapping the PF5102 for a 2sk209 (I used a lsk170B in the spice model), adding a treble control, and 0.47uf input capacitors with 10k resistors.

Spice model:

(R16 is a trim pot and R3 + R2 is the bias trim)

The circuit seems to work as intended in Spice, and when testing the voltage at certain points, everything seems to be fine. I am reading 3.1V at the source of the JFET, 9.3V at the drain of the JFET, 12.47V right before R5, 32.2V at the collectors of Q1 and Q2, and the bias circuit seems to be working as intended. I believe all the grounds are wired properly, and the capsule has been tested in another mic. These voltages seem to correlate with the spice model.

Despite this, when I connect the capsule, I am not getting any signal from the mic. There is no noise or hiss, just a completely silent signal. If anyone has anything I can check for or need to post more information regarding the build, please let me know. Thank you very much, and have a great day.

-Sky

 

[Khron]

Photos of your build?

 

[Voyager10]

Check for shorts between Q1 and Q2's emitters, or Q1/Q2's bases?

 

[skyy7]

Photos of your build?

Its a bit sloppier than I'd like but i think everything is wired correctly.

 

[skyy7]

Check for shorts between Q1 and Q2's emitters, or Q1/Q2's bases?

There don't seem to be any shorts.

Correction: I don't seem to have continuity between the emitter pins, but is that not by design?

 

[Khron]

Correction: I don't seem to have continuity between the emitter pins, but is that not by design?

That's a good thing, they should NOT have continuity to each other.

I don't suppose you have access to an oscilloscope..? Even an old analog one will do.

 

[skyy7]

Check for shorts between Q1 and Q2's emitters, or Q1/Q2's bases?

That's a good thing, they should NOT have continuity to each other.

I don't suppose you have access to an oscilloscope..? Even an old analog one will do.

I wish... I was going to purchase one but couldnt find an old analog one and the digital ones were a touch too expensive...

Would that be the only way to diagnose the problem?

 

[Khron]

Would that be the only way to diagnose the problem?

It would be the easiest. Although alternately, if you're willing to macgyver yourself some sort of probing solution, you could use the instrument input on your interface (assuming you have one) as an AC-coupled oscilloscope input. Which is all you really need, really, just to see if there's any waveform of any sort, at various points in the circuit.

For comparative amplitudes, since you already have the schematic drawn up in LTspice, you could disconnect the ground end of that 50pF capacitor, and insert a sine voltage source between that and ground, and do a transient sim (10-100ms should be plenty).

 

[skyy7]

It would be the easiest. Although alternately, if you're willing to macgyver yourself some sort of probing solution, you could use the instrument input on your interface (assuming you have one) as an AC-coupled oscilloscope input. Which is all you really need, really, just to see if there's any waveform of any sort, at various points in the circuit.

For comparative amplitudes, since you already have the schematic drawn up in LTspice, you could disconnect the ground end of that 50pF capacitor, and insert a sine voltage source between that and ground, and do a transient sim (10-100ms should be plenty).

Oh, great idea. I remember doing this for the hairball comp I built before.

I'll compare it to the Spice model and get back to you. Thank you for the help.

 

[Khron]

One thing to keep in mind is that the optimal bias setting you might have found in simulation will likely NOT be the same that is required in the physical circuit, mostly due to JFET manufacturing tolerances, which are on the "generous" side, to say the very least.

So even before improvising anything else, run that JFET gate bias trimpot from one end to the other and see if you can get any sort of noise or signal out of the mic, at any point along its travel.

 

[skyy7]

One thing to keep in mind is that the optimal bias setting you might have found in simulation will likely NOT be the same that is required in the physical circuit, mostly due to JFET manufacturing tolerances, which are on the "generous" side, to say the very least.

So even before improvising anything else, run that JFET gate bias trimpot from one end to the other and see if you can get any sort of noise or signal out of the mic, at any point along its travel.

No signal, no matter what the pot is set to. Although I think it is biased correctly (I don't have an oscilloscope, but I believe I read that I can get the voltage before R5, divided by 4, and that should be the voltage at the source).

Regarding the AC test signal out of my interface, I seem to be getting 0.084 VAC out of the drain and the source of the 2SK209. I also seem to get 0.084 VAC before and after C4. After R14, it turns to 0 VAC, which I assume is my problem.

The 0.084 VAC reading at the drain and source also seems to be a bit strange, as I am getting 0.18 VAC on spice.

When injecting a larger 2 VAC into the circuit, I still get 0VAC across the 10K resistor.

However, when I inject a 0.1 VAC signal before C4, I get a small output, immeasurable by my voltmeter, but that is present in my DAW.

 

[Khron]

What if you remove R16 & C2 from the circuit?

 

[skyy7]

R2 and R3 is the bias trim pot. I'll try removing the eq circuit.

Update: I'm stupid... sorry i read that as R2 instead of C2... I am removing the eq circuit right now.

 

[JohnRoberts]

A useful method when troubleshooting a circuit is to measure node voltages with a VOM and note them on the schematic.

Voltages that do not agree with typical device operation can suggest what to check.

JR

 

[skyy7]

What if you remove R16 & C2 from the circuit?

I'm still getting nothing.

 

[skyy7]

A useful method when troubleshooting a circuit is to measure node voltages with a VOM and note them on the schematic.

Voltages that do not agree with typical device operation can suggest what to check.

JR

I guess based on this, it could be the JFET since I'm not getting enough VAC out of the source and drain but I've tried an LSK170 in the circuit and used multiple 2sk209s... I ordered some J113s from mouser for 1-day delivery yesterday, but I feel like this circuit should work with a 2SK209 based on the specs sheets.

 

[skyy7]

A useful method when troubleshooting a circuit is to measure node voltages with a VOM and note them on the schematic.

Voltages that do not agree with typical device operation can suggest what to check.

JR

Oh, never mind. I removed R14 and R15 and it seems to work now.... The resistors are working, though. Is there an issue with how I designed the circuit? R14 and R15 were not in the original circuit, but I assumed it would work as I saw others add it to schoeps style circuits. It also worked fine in Spice.

 

[skyy7]

https://audioimprov.com/AudioImprov/Mics/Entries/2014/2/8_EQ_Pt.3_-_Transformerless_Mics.html
This link is where I based the EQ circuit.

 

[Khron]

I removed R14 and R15 and it seems to work now....

Removed, as in, replaced / bypassed them with a couple of wires, or...?

 

[skyy7]

Removed, as in, replaced / bypassed them with a couple of wires, or...?

Bypassed