Comparison of JFETs for mic applications

[k brown]

Not many applications use T0-92 now. Shouldn't there be a small smt equivalent available? Do those small capsules with a jfet built in use a bare die? I always assumed they had a SOT package inside, but I never disassembled one to check.

What full size mic's circuit board doesn't have room for a TO-92?

Here are the FETs in TSB-140A, Panasonic WM-61 and Primo EM172 (all look like same pkg):

 

[thor.zmt]

Very interesting circuit! How are you supplying capsule polarization voltage?

Any way you like. If you ask me, to the backplate.

Thor

 

[thor.zmt]

So why does it turn up in so many mics?

China Syndrome.

Original work is considered bad. Companies refuse to pay competent designers for competent designs, if they can copy something that "works".

Or worse, they just buy in all parts and than assemble a Mic, usually a worse one than one would DIY.

And Western "Brands" buy these unchanged and put their names on them and advertise and hype the proverbial out of these mic's, because they caught the China Syndrome too and cannot see why they should pay a competent engineer, who costs money, when they can just get the China factory to put a green paint job into their mike.

Thor

 

[soliloqueen]

China Syndrome.

Original work is considered bad. Companies refuse to pay competent designers for competent designs, if they can copy something that "works".

Or worse, they just buy in all parts and than assemble a Mic, usually a worse one than one would DIY.

And Western "Brands" buy these unchanged and put their names on them and advertise and hype the proverbial out of these mic's, be cause they caught the China Syndrome too and cannot see why they should pay a competent engineer, who costs money, when they can just get the china factory to put a green paint job into their mike.

Thor

in my experience, it's also related to licensing and regulatory costs. these vintage designs don't usually have RF oscillators, so they don't need to be licensed or third party tested. modern chinese designs had their licensing status established decades ago, so that's also cheaper. new performant designs have to be licensed both federally and, depending on the power supply configuration, also locally.

 

[ccaudle]

What full size mic's circuit board doesn't have room for a TO-92?

It isn't so much space as what is cheapest to manufacture in current factory processes.

FETs in TSB-140A, Panasonic WM-61 and Primo EM172

The Panasonic and Primo capsules are discontinued. Do the current replacements still have what appears to be TO packages that required manual lead bending, or have they changed to standard machine placeable smt packages?

 

[thor.zmt]

new performant designs have to be licensed both federally and, depending on the power supply configuration, also locally.

A lot of factories just put the FCC mark on and leave it to the customer to deal with the FCC.

But the key fault is the chinese education system. Repetition and copying are emphasised. Few products of this system (including engineers) are capable of any original work.

Something like the basic tricks I applied to the circuit to make it a charge amp are beyond MOST Chinese EE's, certainly most I encountered in person.

And the reason they do not copy schoeps is that as you say it includes an oscillator and stepup to the Bias. But the FCC is not the issue here. The issue is they cannot copy the existing design.

The schematics on line (or in Chinese publications - I used to have a shelf full of annual magazine books full of schematics of western and Chinese audio gear) are useless to chinese EE's because they cannot buy a part labelled Dr1 or Dr2.



Not being used to german style drawings they will not grock that these are chokes/inductors (and expecting them to actually understand the circuit and figure they are chokes and derive suitable values is far beyond most of them) and so operation "copy schoeps" fails.

The realisation that you can probably optimise the circuit for P48 and derive the capsule bias from this is another bridge or a million too far for them.

And yes, I am ABSOLUTELY SERIOUS HERE.

Thor

 

[k brown]

It isn't so much space as what is cheapest to manufacture in current factory processes.



The Panasonic and Primo capsules are discontinued. Do the current replacements still have what appears to be TO packages that required manual lead bending, or have they changed to standard machine placeable smt packages?

Current Primo product is EM272/273.

They look the same as the 172, but I have no idea what FET is inside; I'm trying to figure why it matters . . .

I thought this thread was about FETs for DIY condenser mics.

 

[jp8]

2SK879 and 2SK880, abundantly available from Mouser and Digikey

 

[soliloqueen]

interested in seeing an LS846 in the u87 style circuit

 

[Voyager10]

Very interesting circuit! How are you supplying capsule polarization voltage?

A full version of the circuit would take a voltage from XLR pins 2 and 3, via some high-value (> 100K) resistors, through one or two RC filters. The voltage you get depends on the current drawn by the amp, but will typically be around 40V.

Note that the "ADK" circuit is intended to drive a low-ratio (typically 2:1 or 1.85:1) output transformer, not directly to the XLR pin. This circuit is just a test fixture I'm using to look at the FET amplifier stage - my preamp has a 3K input impedance so it's quite easy to drive.

 

[Voyager10]

Ali, as well, but you suspect they are fakes?

The main problem here is whether my results are reproducible by others. If FET X from eBay/Ali measures badly, does that mean all FET X's are bad, or have I just got a fake / reject sample? (This happened with the 2SK170, for instance).

Conversely, if I post 'FET Y is wonderful', that doesn't help anyone if the only sources for FET Y are unreliable. The very last thing I want to do here is create more mythology around now-obsolete devices.

 

[micolas]

The main problem here is whether my results are reproducible by others. If FET X from eBay/Ali measures badly, does that mean all FET X's are bad, or have I just got a fake / reject sample? (This happened with the 2SK170, for instance).

Conversely, if I post 'FET Y is wonderful', that doesn't help anyone if the only sources for FET Y are unreliable. The very last thing I want to do here is create more mythology around now-obsolete devices.

It is perfectly justified what you want to highlight!

I apologize if I deviate a little from the central topic of the thread
Now, something I reported in another collateral post (mea culpa, I don't know where, that's why we have to keep the thread theme):
Interesting and encouraging:
Being driven by curiosity, I bought a substantial batch of 2sk30a from Ali last year, I will look for the seller's link, although I suspect that most of them are from the same manufacturer. Nor was I under any illusion that I was really what the label said.
I tested them all, compared to the Toshiba ones, in a Fame Studio C01 microphone, (jFet+BJT+Transf.) Also, I fully measured the Ali batch for Idss
It was an unexpected pleasant surprise. The entire Ali batch was very consistent and identical to my old genuine Toshibas.
And some Alis had superior behavior, less inherent noise and a sound that I preferred!
Now, I'm tempted to try with other jFETs from Ali (The resistors with the metal film and the zenner diodes also had good quality, but regarding the more demanding capacitors I have doubts)
Maybe others have also had positive experiences with active electronic components from Ali.

 

[GJB17]

Hi people... I'm noticing a lot of recent discussion about FET types, availability, specs, substitutions, counterfeit and rebadged or no-name items, etc... etc.
As has been noted by others (e.g. post 69)... I still have a large "stash'" of genuine JFETs that were recovered from a local company that closed back in early 2,000's some time. They are GENUINE... as the feedback here, on Ebay and others reflects. the stock is here with me in Perth, Western Australia... not in a shed in QuonDong... or wherever!

FYI... a list below. I like to sell in at least 50 piece lots... 100 preferred.
I also have a large quantity of 2SA and 2SC (audio type) BJT transistors.

Cheers,
Graham

2SJ103BL

2SK117-BL

2SK117-GR

2SK246-BL

2SK30ATM-GR

2SK30ATM-Y

2SK369-BL

2SK369-V

2SK373-GR

2SK709-V

 

[OneRoomStudio]

@Voyager10 have you had the chance to test any of the Linear Systems parts? It would be interesting to see how LSK170X or LSK189 stack up.

 

[thor.zmt]

Maybe others have also had positive experiences with active electronic components from Ali.

Being in Thailand, I buy shedloads of Ali.

Knowing how to deal with chinese vendors and how to spot fakes, I get burned around 1 in 4 times, which I consider acceptable, but highly annoying.

Thor

 

[micolas]

Being in Thailand, I buy shedloads of Ali.

Knowing how to deal with chinese vendors and how to spot fakes, I get burned around 1 in 4 times, which I consider acceptable, but highly annoying.

Thor

Thor you are a mine of precious resources, of valuable information!

 

[rmburrow]

What might be illuminating here is to include some of the FET's parameters, such as Ciss (Gate to Source capacitance) and Crss (Reverse transfer Capacitance Drain to gate) as well transconductance and IF POSSIBLE a proxy for transconductance linearity with signal.

For now with casual glances, it seems that in this circuit the "Key Figure Of Merit" (KFOM) is Crss.

Thor

Has anyone tried to measure the input impedance (or input admittance) for some of these fet's? I assume the impedance would be some ultra high resistance in parallel with a few pf of capacitance. Hopefully the input capacitance (reactance) of the fet is negligible compared to the 20 to 40 pf of the capsule in parallel.

 

[MicUlli]

@Voyager10 have you had the chance to test any of the Linear Systems parts? It would be interesting to see how LSK170X or LSK189 stack up.

I have tested the LSK170A regarding noise. Dont waste your money

 

[MicUlli]

Has anyone tried to measure the input impedance (or input admittance) for some of these fet's? I assume the impedance would be some ultra high resistance in parallel with a few pf of capacitance. Hopefully the input capacitance (reactance) of the fet is negligible compared to the 20 to 40 pf of the capsule in parallel.

With bootstrapped JFETs Cin can be neglected, but in charge amps it plays a significant role..
Reading the data sheets can help.

 

[ricardo]

Has anyone tried to measure the input impedance (or input admittance) for some of these fet's? I assume the impedance would be some ultra high resistance in parallel with a few pf of capacitance. Hopefully the input capacitance (reactance) of the fet is negligible compared to the 20 to 40 pf of the capsule in parallel.

Just look at the Specs.

eg 2sk170 has Ciss 30p and Crss 6p.

That's why it isn't good for SDCs and only OK (but bla) for LDCs and only in certain circuits. There are other reasons why it isn't a good choice for condensor mikes.