Reply by October 20, 20172017-10-20
On Tuesday, 18 July 2017 14:29:21 UTC+1, Steve M  wrote:
> I'm looking for a replacement for the CA3039 diode array, which has been successfully used in a precision demodulator circuit operating up to 1 MHz. The part has been obsoleted by Intersil. The important characteristics are low leakage current (less than 0.1 nA at 25C with 1 V reverse bias) and close matching of forward voltage drops. Also, the forward voltage drop with 1 nA or less current needs to be 200 mV or greater. Most common diodes turn on much earlier. The operating range is from 20C to 60C. > Thank you for your help.
I have a spare new CA3039 if you are interested?
Reply by Tauno Voipio July 19, 20172017-07-19
On 19.7.17 19:07, Winfield Hill wrote:
> xmacleod@gmail.com wrote... >> >> Thank you for the comments and suggestions. I made some >> measurements of a diode connected transistor, 2N3904 BC >> shorted, and it provides the low leakage and forward >> voltage characteristics needed. Interestingly, the forward >> voltage characteristics are more like the CA3039 than a >> switching diode. For instance, the forward voltage at >> 0.1 uA is about 400 mV, whereas for a 1N4148 diode, it >> is about 150 mV. Also, the CA3039 diode array has a typical >> reverse breakdown of 5 V, similar to a transistor BE junction. >> Is the CA3039 actually an array of diode connected transistors? >> I will look further in to using matched transistor arrays. >> Any recommendations for devices are welcome. > > The high voltage you refer to means the current density > is higher, which means the part is smaller. Small dies > also have lower capacitance. I suppose that's actually > the parameter you're concerned with? Working at high > wold normally imply approaching distortion.
My guess is that the OP is using a ring mixer. Its output level is limited by the forward drop of the diodes. It seems thet this was the reason the first ring mixers (modulators) in telephony carrier system were made with several diodes in series in each branch. -- -TV
Reply by Winfield Hill July 19, 20172017-07-19
xmacleod@gmail.com wrote...
> > Thank you for the comments and suggestions. I made some > measurements of a diode connected transistor, 2N3904 BC > shorted, and it provides the low leakage and forward > voltage characteristics needed. Interestingly, the forward > voltage characteristics are more like the CA3039 than a > switching diode. For instance, the forward voltage at > 0.1 uA is about 400 mV, whereas for a 1N4148 diode, it > is about 150 mV. Also, the CA3039 diode array has a typical > reverse breakdown of 5 V, similar to a transistor BE junction. > Is the CA3039 actually an array of diode connected transistors? > I will look further in to using matched transistor arrays. > Any recommendations for devices are welcome.
The high voltage you refer to means the current density is higher, which means the part is smaller. Small dies also have lower capacitance. I suppose that's actually the parameter you're concerned with? Working at high wold normally imply approaching distortion. -- Thanks, - Win
Reply by George Herold July 19, 20172017-07-19
On Wednesday, July 19, 2017 at 9:05:34 AM UTC-4, xmac...@gmail.com wrote:
> Thank you for the comments and suggestions. I made some measurements of a diode connected transistor, 2N3904 BC shorted, and it provides the low leakage and forward voltage characteristics needed. Interestingly, the forward voltage characteristics are more like the CA3039 than a switching diode. For instance, the forward voltage at 0.1 uA is about 400 mV, whereas for a 1N4148 diode, it is about 150 mV. Also, the CA3039 diode array has a typical reverse breakdown of 5 V, similar to a transistor BE junction. Is the CA3039 actually an array of diode connected transistors? > I will look further in to using matched transistor arrays. Any recommendations for devices are welcome. > Thanks
I made (and sell) a 'silly' audio mixer that is an audio transformer and diode ring. I found that I got better performance out of diode connected transistors than from other diodes I tried. George H.
Reply by July 19, 20172017-07-19
Thank you for the comments and suggestions.  I made some measurements of a diode connected transistor, 2N3904 BC shorted, and it provides the low leakage and forward voltage characteristics needed. Interestingly, the forward voltage characteristics are more like the CA3039 than a switching diode.  For instance, the forward voltage at 0.1 uA is about 400 mV, whereas for a 1N4148 diode, it is about 150 mV. Also, the CA3039 diode array has a typical reverse breakdown of 5 V, similar to a transistor BE junction.  Is the CA3039 actually an array of diode connected transistors?  
I will look further in to using matched transistor arrays.  Any recommendations for devices are welcome.
Thanks       
Reply by whit3rd July 19, 20172017-07-19
On Tuesday, July 18, 2017 at 6:29:21 AM UTC-7, xmac...@gmail.com wrote:
> I'm looking for a replacement for the CA3039 diode array, which has been successfully used in a precision demodulator circuit operating up to 1 MHz. The part has been obsoleted by Intersil. The important characteristics are low leakage current (less than 0.1 nA at 25C with 1 V reverse bias) and close matching of forward voltage drops. Also, the forward voltage drop with 1 nA or less current needs to be 200 mV or greater. Most common diodes turn on much earlier. The operating range is from 20C to 60C.
Presumably you like matching for the same reason RF duals for PIN switches or protection are matched. Would a trio of matched duals do as well as a set of six? <http://www.skyworksinc.com/uploads/documents/SMP1330_Series_200050N.pdf> The usual diode arrays are, nowadays, not tested for any kind of matching. If you want enough, though, a manufacturer can test for any criteria you care to specify. The specs above are NOT what a CA3039 datasheet shows, but can be inferred from other data in the sheet.
Reply by July 18, 20172017-07-18
>I'm looking for a replacement for the &nbsp;CA3039 diode array, >which has been successfully used in a precision >demodulator circuit operating up to 1 MHz. The part has been >obsoleted by Intersil. &nbsp;The important characteristics >are low leakage current (less than 0.1 nA at 25C with 1 V >reverse bias) and close matching of forward voltage drops. >Also, the forward voltage drop with 1 nA or less current >needs to be 200 mV or greater. &nbsp;Most common diodes turn >on much earlier. The operating range is from 20C to 60C. &nbsp;
It's probably time to replace the old-school method with a nice fast CMOS mux. You can get the desired dead zone by controlling the duty cycle. A simple method is to use a 74HC4017 1-of-10 decade counter: OR together outputs 1-4 for one switch phase, and 6-9 for the other. You'll get a one-clock deadband at the switching points. Cheers Phil "also from the old school" Hobbs
Reply by Jeff Liebermann July 18, 20172017-07-18
On Tue, 18 Jul 2017 06:29:14 -0700 (PDT), xmacleod@gmail.com wrote:

>I'm looking for a replacement for the CA3039 diode array, which has been successfully used in a precision demodulator circuit operating up to 1 MHz. The part has been obsoleted by Intersil. The important characteristics are low leakage current (less than 0.1 nA at 25C with 1 V reverse bias) and close matching of forward voltage drops. Also, the forward voltage drop with 1 nA or less current needs to be 200 mV or greater. Most common diodes turn on much earlier. The operating range is from 20C to 60C. >Thank you for your help.
Have you considered buying the real part from a surplus parts dealer? You probably won't like their prices, but it will be less than the cost of redesigning the device, approving a new vendor, inspecting parts, etc. <http://www.integrated-circuit.com/products-detail/CA3039/EIS-398091.html> <http://www.key-chips.com/products-detail/4307116/CA3039.html> <http://www.4starelectronics.com/part_detail/CA3039.html> etc... -- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
Reply by George Herold July 18, 20172017-07-18
On Tuesday, July 18, 2017 at 9:29:21 AM UTC-4, xmac...@gmail.com wrote:
> I'm looking for a replacement for the CA3039 diode array, which has been successfully used in a precision demodulator circuit operating up to 1 MHz. The part has been obsoleted by Intersil. The important characteristics are low leakage current (less than 0.1 nA at 25C with 1 V reverse bias) and close matching of forward voltage drops. Also, the forward voltage drop with 1 nA or less current needs to be 200 mV or greater. Most common diodes turn on much earlier. The operating range is from 20C to 60C. > Thank you for your help.
How about a quad transistor array? Use the BE or BC diode... I see matched ones at Digikey, but spendy. George H.
Reply by John S July 18, 20172017-07-18
On 7/18/2017 8:29 AM, xmacleod@gmail.com wrote:
> I'm looking for a replacement for the CA3039 diode array, which has > been successfully used in a precision demodulator circuit operating > up to 1 MHz. The part has been obsoleted by Intersil. The important > characteristics are low leakage current (less than 0.1 nA at 25C with > 1 V reverse bias) and close matching of forward voltage drops. Also, > the forward voltage drop with 1 nA or less current needs to be 200 mV > or greater. Most common diodes turn on much earlier. The operating > range is from 20C to 60C. Thank you for your help. >
You must have measured the above parameters yourself. The datasheet does not even get close to your requirements. How do you expect people to help you? I don't think anyone will purchase a bunch of diodes and parametertize them for you.