Millennium bypass - J-Fets and diodes

[Lino22]

Guys what commonly available J-Fets would you recommend for Millennium Bypass? I can't find any reasonably priced J201.
Would you recommend any substitute available in EU? What parameter shall i look for?

I used J112, and it was on all the time - i mean the LED was bright and then less bright. There was 13mA going through the S-D branch. Shall i just raise the current limiting resistor R3 in the S-D branch?

Also, what common diodes have smaller reverse current than 1N4148? I would like a slower raise time of the LED.

[antonis]

Use MosFet (2N700 or BS170) for Millenium 2 and don't bother with those        JFETs anymore..

[Lino22]

I will, but i have had a bunch of PCBs made. I would like to use them 

[duck_arse]

use a BAT series schottky diode for slow turn ons. == well, in a proper mill, yes, but I dunno about in a jfet mill.

[Clint Eastwood]

Guys what commonly available J-Fets would you recommend for Millennium Bypass? I can't find any reasonably priced J201.
Would you recommend any substitute available in EU? What parameter shall i look for?

I used J112, and it was on all the time - i mean the LED was bright and then less bright. There was 13mA going through the S-D branch. Shall i just raise the current limiting resistor R3 in the S-D branch?

Also, what common diodes have smaller reverse current than 1N4148? I would like a slower raise time of the LED.

You are not using 9V like in the original circuit but 18V, that mainly causes the high current draw. I tried in the simulator, and raising R3 to 2.7k would result in about 5mA current draw when on, and 15uA when off.
There is nothing wrong with jfets, but I guess Antonis just does not have the patience to work with them

[antonis]

There is nothing wrong with jfets, but I guess Antonis just does not have the patience to work with them

Gate Cut Off Voltage, Shorted Gate Drain Current, Transconductance, Dynamic Output Resistance, Amplification Factor and all other parameters great uncertainity call for either patience of Job or rather complicated configurations..

[Rob Strand]

e Cut Off Voltage, Shorted Gate Drain Current, Transconductance, Dynamic Output Resistance, Amplification Factor and all other parameters great uncertainity call for either patience of Job or rather complicated configurations..

On the other side of the coin, people have had issues getting a good balance of leakage currents from the diodes with the Millenium 2.   I suspect PCB leakage might be screwing things up.  Some people used Schottky's to compensate but that does push the diode leakage currents up quite a bit (and hence the risk of switching tics if the *DC* output impedance of the pedal isn't low.)

FYI, way back in 2002 I did a transistor only version; the aim was to come up with a FET/MOSFET-less circuit.   It's not as good as the JFET Millenium I but it might go up against a hacked Millenium II.   IIRC some finer points are discussed in the thread,
https://www.diystompboxes.com/smfforum/index.php?topic=122573.msg1156942#msg1156942

[merlinb]

I will, but i have had a bunch of PCBs made. I would like to use them 

So put a MOFET in the JFET spot, you only have to crossover a pair of legs to get the pinout right, it works the same otherwise. You don't actually need two diodes like in RG's example.

[Lino22]

Guys thanks for your advices. I made a little testing apparatus and made it work with 10k restricting resistor and J112. It has 1.5mA current draw with 18V DC.
The next step will be testing diodes to make it start up more smoothly - Millennium really helps with a turn/on pop of hi gain devices when you have everything full blast (no matter if the input is grounded/pulled-down or not).

[R.G.]

Many of the facets of this were covered in the series of articles on the Millenium on geofex.com, including the dim glow sometimes.

Quick comment on numbers of diodes and semiconductor leakage.  How many diodes you need relies on imponderables like what the various diode leakages really are. In general, semiconductor makers specify a maximum leakage current, not a minimum, so leakages down into femto-amps are just fine by the manufacturers. They pass the no-bigger-than test.
And there is a long progression of silicon devices getting better and better over time as better-performing fabs come on line and the just-past-generation semiconductor fabs are relegated to jellybean device manufacture or sold off to other players in the semiconductor market. As a result, you often can't get devices as leaky as you used to. Real 1N914s and similar had to be doped with gold atoms to get them to quench the stored charge in the depletion zone fast enough to be "fast switching". Devices like the 1N4148 and similar managed to get the same or faster switching without the gold doping that caused the high leakage of the older 1N914.
Then some clever semi maker figured out that their new 1N4148's passed the specs for 1N914s and that they could charge the gotta-have-the-real-part-number market extra for vintage part numbers without actually doing that gold doping. Hmmmm...
Real 1N4148s and other switching types also continue to (probably) get better as they are moved to newer semiconductor processes and finer geometry in the fabs. So just like "1N914" may not imply "gold doped and higher leakage", a 1N4148 may have just under the datasheet maximum leakage, or may be dramatically less leakage.
Bottom line: be prepared to sub in some different devices if you're going to rely on no-minimum or no-maximum specs on a datasheet. It's even worse if you have only "typical" specs. You'll nearly always find some vintage of parts from the same maker and some other maker that will come in acting a way you don't expect.
The migration of semiconductors from older to newer processes has been going on for the last 50-60 years, and for 24 years since the Millenium articles were first written. Expect variations, and be pleasantly surprised if they don't happen. Especially where you're doing the difference between a high-leakage and a low-leakage part. Adding more paralleled "leaky" parts may be needed. If you got leaky ones, the low-leak pull down might not be needed.
And as John Wyndham and Paul Kantner remind us, life is change -that's how it differs from the rocks.   8-)

[PRR]

articles on the Millenium on geofex.com,

This fails on my PC. It may be browser or O/S dependent. AFAICT your current host does not support httpS and/or my browser/OS won't fall-back to non-S.

This works for me, today: http://geofex.com/

[Rob Strand]

I get two fails.  https vs http related.

https://geofex.com/

Code Select Expand

Secure Connection Failed

An error occurred during a connection to geofex.com. PR_CONNECT_RESET_ERROR

Error code: PR_CONNECT_RESET_ERROR

    The page you are trying to view cannot be shown because the authenticity of the received data could not be verified.
    Please contact the website owners to inform them of this problem.

Learn more...

http://geofex.com/

Code Select Expand


HTTPS-Only Mode Alert
Secure Site Not Available

You've enabled HTTPS-Only Mode for enhanced security, and a HTTPS version of geofex.com is not available.

Learn More...
What could be causing this?

    Most likely, the website simply does not support HTTPS.
    It's also possible that an attacker is involved. If you decide to visit the website, you should not enter any sensitive information like passwords, emails, or credit card details.

If you continue, HTTPS-Only Mode will be turned off temporarily for this site.


[R.G.]

Geofex.com was relocated to a new host, and apparently the https is still not hooked up. These are the direct links, but for an https-rejecting browser, they probably will throw an error too.

http://www.geofex.com/article_folders/millenium/millen.htm
http://www.geofex.com/article_folders/mill2extn/mil2plus.htm
http://www.geofex.com/Article_Folders/Millenium/The%20Next%20Millenium.pdf
http://geofex.com/FX_images/Millenium%20Wiring.pdf

[Rob Strand]

Geofex.com was relocated to a new host, and apparently the https is still not hooked up. These are the direct links, but for an https-rejecting browser, they probably will throw an error too.

http://www.geofex.com/article_folders/millenium/millen.htm
http://www.geofex.com/article_folders/mill2extn/mil2plus.htm
http://www.geofex.com/Article_Folders/Millenium/The%20Next%20Millenium.pdf
http://geofex.com/FX_images/Millenium%20Wiring.pdf

They work without any errors.   Just enough to get under the radar.

[PRR]

These are the direct links, but for an https-rejecting browser, they probably will throw an error too.

They work for me in the everyday FF browser but I recall hacking its httpS setting when that became an issue.

Ah, I recently installed Floorp browser, yawned, but didn't uninstall it. It renders all the URLs in R.G.'s post #12. ---- Ah, my Floorp defaulted to "Don't enable HTTPS-Only Mode". If I "Enable HTTPS-Only Mode in all windows", I get
HTTPS-Only Mode Alert
Secure Site Not Available
You've enabled HTTPS-Only Mode for enhanced security, and a HTTPS version of www.geofex.com is not available.
    Most likely, the website simply does not support HTTPS.