ECM8000 Mods

Yes, I sent the files to Flatpicker, who kindly agreed to host them for me. I recorded them through my custom built (of course by me) tube transformer coupled pre and Canare cables straight to Alesis Masterlink. For recording I used 24bit/96K. However, after converting files to MP3, the difference between mics is less apparent. Anyway, I want you guys to listen to it and deside which one is a stock and which is a modded one, if you can. Also, I will send the mics to Harvey for complete evaluation.
 
Thank you Flatpicker!

OK Gentlemen,

We have two files--Left and Right. Now please take your pick-- which one is a modded and which is a stock one? Or both of them sound the same? Please speak, but not all at the same time.

Your nice comments about my new pre will be also greatly appreciated :)
 
Something struck me recently that you guys might think about trying: if the FET used in a capsule has lithographic symmetery between the source and drain at the IC level, which is not altogether uncommon with FET's, then the 2-wire linkwitz modified capsule is _identical_ to the original capsule with the leads swapped.

A quick experiment done by one of the guys on the micbuilders Yahoo email list shows that the gain is nearly the same both ways with the same load resistor. The gain in common source mode was a good deal higher.

To the extent that this symmetry holds, a two wire capsule can be put in source follower mode just by reversing it in the circuit it comes with.


Bob
 
the difference is diminished

Yep, the difference really is less pronounced on my two mediocre monitoring chains listening to .mp3's as it was with Marik's high-quality monitoring set-up and high-resolution audio. But I hear one clearly better than the other in frequency extension and a little bit in depth, or resolution, when I turn the clips up loud. One has less of the slight nasal quality of the other to my ears.

It really is a nice pre-amp. The playing was very light and quiet--much quieter than the level where most people will play it back.

There is some apparent white noise in the fade out that is actually my arm sweeping the guitar top. Hey, I'm a hack.
 
<if the FET used in a capsule has lithographic symmetery between the source and drain at the IC level, which is not altogether uncommon with FET's, then the 2-wire linkwitz modified capsule is _identical_ to the original capsule with the leads swapped....To the extent that this symmetry holds, a two wire capsule can be put in source follower mode just by reversing it in the circuit it comes with.>

Bob, thank you for this information. I never tried 2 wire Linkwitz, and for ECM8000 application I don't see how to implement it easily because of given polarity of PSU. It is an interesting idea--usually you can use FET's drain and source reversed, with only difference of their capacitance. Here, another difference is that, as you mentioned before, there is an internal diode between gate and source... I just did not think about this arrangement.
 
Marik said:

Bob, thank you for this information. I never tried 2 wire Linkwitz, and for ECM8000 application I don't see how to implement it easily because of given polarity of PSU. It is an interesting idea--usually you can use FET's drain and source reversed, with only difference of their capacitance. Here, another difference is that, as you mentioned before, there is an internal diode between gate and source... I just did not think about this arrangement.
Click to expand...

If you try it, please report how well it works for you. I came up with the idea just considering the topology of the circuit and the device and someone else on micbuilders gave it a quick checkout.

A good indication that the point is valid would be that the sensitivity would go down by 6 dB or so after flipping it but that's not conclusive.

What would be conclusive is an experiment that compared the sensitivity with the two wire mod and supply reversed with just flipping the capsule and using the same load resistor in both cases. That's what the fellow on micbuilders tried and the sensitivities were not quite identical but substantially the same.

Another good indication would be that with a flipped capsule the sensitivity would not vary much over a farily wide range of load resistors. It will in common source mode.

Recent investigations are that the diode that was considered integrated is nothing more than the diode formed by the gate and the source and that a regular old FET is used with leakage employed to bring the gate to the same potential as the source. It was this observation that led me to consider flipping the capsule.

So that I can become as famous as Seigfried I'm gonna call this the Cain modification. :-)

Just kidding, just kidding.


Bob
 
<Recent investigations are that the diode that was considered integrated is nothing more than the diode formed by the gate and the source and that a regular old FET is used with leakage employed to bring the gate to the same potential as the source. It was this observation that led me to consider flipping the capsule.>

Bob, that's interesting. After you mentioned it first, I took my Fluke and measured the FET in diode mode. The tester behaved as if there is diode between gate and source--that's when I believed that actually there is an internal diode--otherwise, because of the simmetry of the FET, gate to drain would measure the same.
 
Marik said:

Bob, that's interesting. After you mentioned it first, I took my Fluke and measured the FET in diode mode. The tester behaved as if there is diode between gate and source--that's when I believed that actually there is an internal diode--otherwise, because of the simmetry of the FET, gate to drain would measure the same.
Click to expand...

Hmm, being a JFET it should see a diode between gate and either source or drain regardless of how asymmetric they are. I can't explain why it wouldn't see one from gate to drain. Puzzling.

If there actually were one in the reverse direction from gate to source then I could possibly understand your tester showing that there _wasn't_ a diode there but...


Bob
 
Marik
They both sound very similar - and very good - must be your pre. The one labeled "left" seems to be more defined - noticeable to me particularly in the strum of the Em chord mid-sample. But my recording and monitoring equipment are primitive, and my ears' frequency response is far from the norm.
 
<Hmm, being a JFET it should see a diode between gate and either source or drain regardless of how asymmetric they are. I can't explain why it wouldn't see one from gate to drain. Puzzling.>

Yeah, but hey Bob, who knows, probably there is a diode there ;)
You know, you gave me idea to try to reverse drain and source in my, 3-wire arrangement and see what happens.

<They both sound very similar - and very good - must be your pre. The one labeled "left" seems to be more defined - noticeable to me particularly in the strum of the Em chord mid-sample. But my recording and monitoring equipment are primitive, and my ears' frequency response is far from the norm.>

Yeah Crazydoc,

As I and Nat have mentioned above the difference after converting to MP3 is much less obvious then in my 24/94 setup.
Actually, I think that another reason is that Nat played very quietly, where the head preamps work with very little signals and from the very beginning it reduced the difference, I am used to. With wide spectrum signals and higher levels the difference is much more noticable. Next week we will try to record guitar again with higher volumes, and also make some drum OH tracks.
I have some more ideas about the capsule shape and size, so please reply in the 'Earthworks mic' thread if somebody has that information.
 
Marik said:
You know, you gave me idea to try to reverse drain and source in my, 3-wire arrangement and see what happens.
Click to expand...

That would be a perfect test. If the DC operating point and sensitivity remain essentily the same that would verify the symmetry and justify the flipping aproach.

What must be remembered if the simple flip is attempted is that the can is now the signal and if it makes another connection in any way to circuit ground, the capsule will be shorted. No danger because the load will limit the current but no sound either.


Bob
 
<What must be remembered if the simple flip is attempted is that the can is now the signal and if it makes another connection in any way to circuit ground, the capsule will be shorted. No danger because the load will limit the current but no sound either.>

Nope, the can is isolated from the ground. In fact, it floats, as shown on my schematics.
 
Marik said:
<What must be remembered if the simple flip is attempted is that the can is now the signal and if it makes another connection in any way to circuit ground, the capsule will be shorted. No danger because the load will limit the current but no sound either.>

Nope, the can is isolated from the ground. In fact, it floats, as shown on my schematics.
Click to expand...

My comment was more for people that just attempt the flip in something where the can has a ground path other than the terminal connections.


Bob
 
How in the behumpwitz have I managed to avoid this thread?

I have been interested in WM60 based mics for a few months, but I haven't had a chance to actually try some things until now. I'm waiting for the parts to get here.

I've also ordered some of the cardiod capsules. Even though the frequency response is a far cry from flat, I figure they'll still be useful (namely hi hat on stage).


I first came across these ideas on prosoundweb. http://www.prosoundweb.com/recording/tapeop/buildmic/buildmic_16_1.shtml . In fact, the way they've got it drawn out, you can eliminate the Behringer completely and build a much simpler circuit. I wonder how their circuit compares to what you guys have been coming up with. How does their version look?

BTW, I think right sounds more articulate than left. Left sounds a little on the dull side.
 
You must log in or register to reply here.
Back
Top