3. Diode ring, which provides variable gain, used in analog compressors like the Neve 33609 (I have a clone of the 33609, and I’m very fond of it)
Think about this: if you have a nonlinear device like a diode, then the dynamic resistance changes depending on the operating point. If you modulate the operating point, you’re modulating the dynamic resistance.
Abuse minority carrier lifetime to very suddenly turn from resistive to capacitive just after switching from forward current to reverse bias; use the fact that the current wants to keep flowing to force it to concentrate into another step recovery diode that's about to cut out, in turn making the cut off spike even sharper, and on.
Surprisingly capable for e.g. blasting a FET gate off while tanking the Miller effect gate current needs through sheer power of SRD-based-pulse-shaping.
Because for e.g. GaN and SiC if you have to choose between ZVS and ZCS, you can take ZVS and just furnish a gate pulse that _makes_ the channel remain off as the current drops and the voltage soars. At least if you pull some tricks and make the current commutation loop sufficiently low inductance to keep your transistors from blowing out in self-inflicted overvoltage due to a current that needed to pass too high an inductance in too short a time.
(Total drain charge is sadly fundamental to the channel's existence, and non-ZVS turn-on is unavoidably lossy. A majority carrier device is theoretically capable of just switching off though if you can arrange the structure for extremely low inductance.)
> This topic seems to be broadly misunderstood. It is 100% verified fact by both myself and others (including university researchers) that diode strings can produce more heat (or watt-hours, BTU) from a given solar panel than a bare resistance element.
> The reason I put “gate” in scare quotes in the illustration is that the circuits are not readily composable to implement more complex digital logic...
Any good suggestions on resources talking about building complex digital logic out of something more suitable?
While diodes alone are not suitable for complex logic, they were instrumental on making computers cheaper in the late vacuum tube era. Vacuum tubes have fairly low reliability and short usable life so having too many of them in your computer is really bad for the cost and reliability of your system. Early transistors were not much better. They would get better over time, but cheap, reliable mass produced diodes were available long before transistors got there.
And while diodes alone cannot do it, a system with a few vacuum tubes to provide the gain and driving a whole lot of diodes made a lot of computers possible at price points that vacuum tubes alone could only dream of. An example is the hacker folklore sweetheart LGP-30, of The Story of Mel fame. 113 vacuum tubes driving 1500 diodes made for a computer that was the size of a fridge, weighed 800 pounds, drew 1.5kW and cost $50k (~500k in modern money), which made it pretty much a personal computer for the late 50's.
He mentions diode logic and points out the drawback of the limited output current, but doesn't mention the obvious solution of a transistor in voltage-follower configuration.
I always thought RTL was pretty nifty, and it was used in a lot of early computers. I think it's a lot less fussy of component values than the earlier RTL.
A diode can switch off an AC source when a battery is present: see second circuit in accepted answer, introduced by, "Alternatively, you can probably get away with just using some schottky diodes:"
It has one RC constant when charging and a different RC constant when discharging through the diode.
Why would you want to charge a capacitor slowly when power is applied to the device, but discharge it fast when power is cut? There are various applications for that.
For instance, circuits that control some timed behavior, like holding a CPU chip in a reset state at start up while power stabilizes, and then releasing it. You want that circuit to reset itself quickly if power is lost.
Analog circuits have things like that in them: for instance circuits that mute an audio amplifier on power up for a bunch of milliseconds until a capacitor charges. If the power is cycled, you want that timer to reset itself.
In some early computers, the bootstrap was actually a matrix of diodes where you'd remove a diode to get a one and leave it in for a zero. I had a bunch of these boards sometime in the mid 1970's and found you could program a fully populated board with a 9V battery - basically connect it across a diode in a bit position where you wanted a '1', there would be a small but pretty flash from inside the glass case as a zero turned into a one.
When things like the 74S188 were available, we had so much fun squeezing bootstrap code for PDP11's into 2 of them; 32 words by 16 bits was more than enough (later I got code that would boot five different devices into 256 words).
A favorite of mine and one of the most common ways to generate a pretty high voltage DC. The full wave version pairs well with a center tapped secondary of a resonant transformer.
This is excellent but in typical low voltage scenarios (5V or lower) the 600mV diode voltage drop becomes very significant. Simple diode half wave rectification works fine at 100V, but at 3.3V it breaks down.
If you’re into audio, they can easily be used for distortion. You “clip” the top of the audio wave. Usually in a asymmetrical way, to get more pleasant sounding distortion.
I have used some regular diodes today as a way to lower the input voltage and this case is not covered. A diode might be more effective than a buck converter because all I wanted was to have a 0.7V lower voltage and the converter can not work in this condition. Zener diode can but it dissipates too much heat for high-current application.
48 comments
[ 3.1 ms ] story [ 55.6 ms ] threadHow completely unintuitive.
1. Frequency mixer, used for heterodyning, important in radio, so I hear. https://en.wikipedia.org/wiki/Frequency_mixer
2. Log converter, where the output voltage is proportional to the logarithm of the input voltage. https://electronics.stackexchange.com/questions/374440/log-c...
3. Diode ring, which provides variable gain, used in analog compressors like the Neve 33609 (I have a clone of the 33609, and I’m very fond of it)
Think about this: if you have a nonlinear device like a diode, then the dynamic resistance changes depending on the operating point. If you modulate the operating point, you’re modulating the dynamic resistance.
Abuse minority carrier lifetime to very suddenly turn from resistive to capacitive just after switching from forward current to reverse bias; use the fact that the current wants to keep flowing to force it to concentrate into another step recovery diode that's about to cut out, in turn making the cut off spike even sharper, and on.
Surprisingly capable for e.g. blasting a FET gate off while tanking the Miller effect gate current needs through sheer power of SRD-based-pulse-shaping. Because for e.g. GaN and SiC if you have to choose between ZVS and ZCS, you can take ZVS and just furnish a gate pulse that _makes_ the channel remain off as the current drops and the voltage soars. At least if you pull some tricks and make the current commutation loop sufficiently low inductance to keep your transistors from blowing out in self-inflicted overvoltage due to a current that needed to pass too high an inductance in too short a time. (Total drain charge is sadly fundamental to the channel's existence, and non-ZVS turn-on is unavoidably lossy. A majority carrier device is theoretically capable of just switching off though if you can arrange the structure for extremely low inductance.)
> This topic seems to be broadly misunderstood. It is 100% verified fact by both myself and others (including university researchers) that diode strings can produce more heat (or watt-hours, BTU) from a given solar panel than a bare resistance element.
https://www.youtube.com/watch?v=42XIbHA9Dv0
Any good suggestions on resources talking about building complex digital logic out of something more suitable?
And while diodes alone cannot do it, a system with a few vacuum tubes to provide the gain and driving a whole lot of diodes made a lot of computers possible at price points that vacuum tubes alone could only dream of. An example is the hacker folklore sweetheart LGP-30, of The Story of Mel fame. 113 vacuum tubes driving 1500 diodes made for a computer that was the size of a fridge, weighed 800 pounds, drew 1.5kW and cost $50k (~500k in modern money), which made it pretty much a personal computer for the late 50's.
I always thought RTL was pretty nifty, and it was used in a lot of early computers. I think it's a lot less fussy of component values than the earlier RTL.
https://www.youtube.com/watch?v=jvNNgUl3al0
https://en.wikipedia.org/wiki/Baker_clamp
Flyback diode:
https://en.wikipedia.org/wiki/Flyback_diode
A diode can switch off an AC source when a battery is present: see second circuit in accepted answer, introduced by, "Alternatively, you can probably get away with just using some schottky diodes:"
https://electronics.stackexchange.com/questions/71753/whats-...
Also, diodes can be used to provide a controlled discharge path for capacitors when a device is turned off.
The circuit in this EE StackExchange question shows it:
https://electronics.stackexchange.com/questions/471285/capac...
It has one RC constant when charging and a different RC constant when discharging through the diode.
Why would you want to charge a capacitor slowly when power is applied to the device, but discharge it fast when power is cut? There are various applications for that.
For instance, circuits that control some timed behavior, like holding a CPU chip in a reset state at start up while power stabilizes, and then releasing it. You want that circuit to reset itself quickly if power is lost.
Analog circuits have things like that in them: for instance circuits that mute an audio amplifier on power up for a bunch of milliseconds until a capacitor charges. If the power is cycled, you want that timer to reset itself.
Another application: Log amp: https://en.wikipedia.org/wiki/Log_amplifier
This exploits the diode's characteristic V-I exponential curve in amplifier feedback to produce output proportional to the logarithm of the input.
When things like the 74S188 were available, we had so much fun squeezing bootstrap code for PDP11's into 2 of them; 32 words by 16 bits was more than enough (later I got code that would boot five different devices into 256 words).
https://www.falstad.com/circuit/circuitjs.html
AJH Synth Sonic V Diode Ladder Filter. (IMHO AJH make the best eurorack filters out there..)
https://en.wikipedia.org/wiki/Voltage_multiplier
Diode half-wave rectifier https://www.circuitlab.com/editor/4da864/
Diode full-wave (bridge) rectifier https://www.circuitlab.com/editor/f6ex5x/
Diode turn-off time https://www.circuitlab.com/editor/fwr26m/
LED with resistor biasing https://www.circuitlab.com/editor/z79rqm/
Zener diode voltage reference https://www.circuitlab.com/editor/7f3ndq/
Charge Pump Voltage Doubler https://www.circuitlab.com/editor/24t6h3ypc4e5/
Diode Cascade Voltage Multiplier https://www.circuitlab.com/editor/mh9d8k/
(note: I wrote the simulation engine)
A favorite of mine and one of the most common ways to generate a pretty high voltage DC. The full wave version pairs well with a center tapped secondary of a resonant transformer.
https://www.geocities.ws/diygescorp/diodeoctaveup.gif