Hello All,

After reading another ARF member's comments about the use of Schottky diodes as detectors and doing a bit of research on the subject, found some very interesting info on simple FET "infinite impedance" detector circuits. Was really surprised that Frank512 had the same idea at the same time, so decided to "play" with this concept and start a thread. Had first heard of it in a mid-1950's article concerning the first all transistor car radio Delco made for Cadillac...

Today I built a "prototype" based on an on-line article circuit which used two FET'S---one as a constant current source to eliminate distortion. Used a FM3955A dual FET since I had several of these new in my semi-conductor "junk box".

Used an old polyvaricon tuning cap from childhood in first version and it was noisy---not surprising, though, after being tossed around in a parts drawer for 40+ years. Replaced it with an air-variable unit and circuit seemed to work great with output connected to AUX input of old stereo receiver---using only a couple feet of wire as antenna ! But after disconnecting battery, it continued to work the same as before ! Hmmm... No time left to investigate today...

Any input to this thread welcome...

More to come... Below are a couple pics. Will post schematic when able...

Nice pictures. I built this infinite impedance detector to substitute for a simple crystal detector in a high-Q coil set. The AM fidelity is astounding, like FM on one spanish music station.

http://sound.westhost.com/articles/am-radio.htm

Thanks Macrohenry.

I used the "Two FET infinite impedance detector" circuit by the same Guy from this site: http://www.mds975.co.uk/Content/crystalsets5.html

Added a 1 meg pot in series with the "constant current" FET source resistor and used a FM3955A dual JFET which apparently doesn't work as it should, though it still does detect well even without power ! Have a couple more to try this weekend...

Ultimately, want to replace the detector diode in one of my transistor radios with this circuit to see how it does. Have a Motorola 8X26E standing by for experimental "surgery"...

John

The audio clip in your link sounds amazing.

db gain---yes, that audio does sound great...

Had success today with a different FET ! Only have a few different types in my parts box and most of those are designed for switching. Had two 2N5163's and decided to try one in the same circuit minus the constant-current portion. It worked great ! Actually had enough output to drive my stereo headphones on strong signals with just a couple feet of antenna wire !

Connected the audio output to AUX input of stereo receiver once again and it worked excellent---with the amplification, could actually receive the strongest signals without any antenna whatsoever ! Very impressive...

Then decided to try and replace the detector in the Motorola with this circuit, but it overloaded the audio badly---tried attenuating the output, but result wasn't that great---so for now, that experiment is over...

John

I'm gonna try this one in a ic761 as a diode replacer;

I also would like to try out the "Infinite Impedance Detector" in a few transistor radios but the only problem is finding a JFET;s that will work well at low power supply voltages between 3 - 6 volts DC.so I started researching this issue and it seems like you have to find a FET that has a low VGGS (Off) spec and also a low "Pinch Off spec for operating a JFET;s at low power supply voltages. I would love to hear from others concerning this subject.

I took a chance and decided to purchased a variety of different types of N channel JFET'S on eBay and I knew that most of them would be either
defective and out of spec so I decided that it was time to build a JFET tester so I could sort out the bad and good ones. I own a Heathkit IT-3120
which will test IDSS & gm but was unable to test a very important JFET specification called VGS OFF so I decided that it would be a good time to
build a tester that would also test VGS OFF. I started searching the internet and found a very popular JFET tester that would test VGS OFF and
IDSS so I ended up building it and modified it a bit to suit my needs. Here is a picture of the completed tester and also the schematic if anyone is
interested in building it. Today I tested about 45 J210's and also 45 2N5457 and to my surprise 99 % met the VGS OFF and IDSS manufacturers data
specifications. The next step will be to build and test them in a "Infinite Z Detector circuit" to see how well they will perform.

Looks good, Frank !

Are you going to build the same, very simple receiver circuit I did to try it out ? If so, think you'll be amazed on how well it works & sounds !

John

I think that I will build and install the "Infinite Z Detector" in the Emerson 31P66 parts radio that I purchased a while back. I decided to first built the JFET tester so that I could weed out the bad ones and to also pick out the ones that will most likely work better at 6 volts.

Hi All,

I finally started building the ( Infinate Impedance Detector ) prototype test circuit. I decided to install two potentiometers so I can tweak it better. I believe that the original circuit was designed for 12 volts DC and I will be modifying it to operate at a much lower supply voltages that will be between 3 to 6 volts DC because of the transistor radios that I will be using. I will complete the test circuit probably tomorrow and I should have some interesting results by the end of the week.

Ignoring the fact that there's no supply, there's are a problem with the tester schematic. The 10K/10K divider at the non-inverting input to the op-amp will force the inverting input to one half of the supply. That means that the voltage at the source lead of the FET will be at half the supply voltage when testing for VGSoff. The minimum drain current will then be R3 divided by half of the supply voltage. If we assume a 12 volt supply and R3=10K, then Id will be 600uA which is far more than the specified drain current for VGSoff which is usually 1nA (eg., 2N3819) or 2 nA (eg., MPF102)!

Another potential problem with the divider is that the drain test current depends on the supply voltage. Not a problem with a regulated supply, but not so good if you use a battery supply.

I will work on a circuit that will address these problems...

John

I built this tester so I could see what the VGS (off) voltages would actually look like at the much lower voltages that I would be using. The tester was definitely designed for experimenters that only wanted ball park measurements. I tested my FETS I varied the power supply voltage from 3 to 15 volts and the VGS (off) readings varied only a small amount. If you want more accurate results you could build a circuit like the one below.

Final Testing Results - I finally found some time to try out some JFETS in my ( Infinite Z- Detector ). In my first prototype most of the JFET's overloaded and had stability problems when trying to detect strong local AM stations just like John experienced. I then increased the bypass cap to 1 uf on the + 3 volts going to the drain and immediately the circuit became stable and useable. I then tried my 2N5457, J210 and J305 that I had and I surprised that most of them worked quite well. The 2N5457 produced some distortion at very strong signal levels and was also a little unstable when trying to tune in the strong station but produced very nice sounding audio. The J210 and J305 were the most stable but the J305 sounded better and produced a smoother sound with more bass. I then tried a BAT-46 " Schottky " diode and it produced louder audio then the FET's but it was did not sound as clear as the FET's. I was a little disappointed that the ( Infinite Z- Detector Circuit ) was not too good at detecting weak signals but it did produce nice sounding audio. The next 2 projects will be to build a ( Sensitive AM Envelope Detector ) and I will also will try biasing the BAT-46 to see if this will help with detecting weak signals. Below is the generic schematic for the ( Sensitive AM Envelope Detector ) that I found on the Internet.

Hello Frank,

Interesting that the bypass cap made such a difference. Am interested in your envelope detector results as I have the parts available to build it...

John

It is good to hear from you John. I had pretty long leads coming to and from my radio to the copper clad test board which may have contributed to the stability issues I experienced but after adding the 1 uf it cured most of the feedback oscillations. I will also start building the "Sensitive AM Envelope Detector" circuit that I saw at http://zpostbox.ru/sensitive_envelope_d ... iagram.htm I also found another interesting low distortion transistor AM detector at http://www.pan-tex.net/usr/r/receivers/ ... detect.htm which I think I will also build. Let me know how you make out.

Hello Frank,

Will wait for your test results before building the "envelope detector" myself---want to know that it works before taking the time with all else going on here... Do make time for the radio refurbish projects to maintain sanity, but experiments take more patience than I can come up with most days unless really motivated... Also will wait for your report on the other low distortion circuit...

Glad all is okay and you're able to play with this again...

John

patiently waits for an arfer to roll a hifi am detector board populated with parts for instal into detector diode position....

After searching through the Mouser online catalog I found a really sensitive zero bias Schottky detector diode that' are used in RFID (Radio Frequency ID) tags on various items stores sell. They are SOT-23 which are very small so I soldered it to my SMT prototyping board I had which will make it much easier to handle and to install as a drop-in replacement for my germanium diodes that are in my transistor radios. I will keep everyone updated on how ll this supposedly sensitive diode works in my transistor radio.

Well I tested this new supposedly sensitive HSMS-2850 zero bias Schlocky detector diode and it performed about the same as a germanium diode.