Photo Fet like H11F1 with higher signal handling?

[MarshallPlexi]

I've been studying the H11F1 all night and was getting really excited to try it since I thought it would work nicely as a vactrol replacement. I have an inherent mistrust in vactrols since I fear they will be discontinued shortly and seem to have widely varying tolerances.

Basically what I need is a voltage controlled resistor for various uses.

Anyone know of something like the H11F1?

[NFX]

TLP222G possibly

[PRR]

The output linearity is inversely related to sensitivity (how hard you have to hit the LED to make it turn on).

Because 99.44% of applications want easy turn-on and do not care about linearity, you are not likely to find anything "better".

[MarshallPlexi]

TLP222G possibly

Tried the TLP222G and it was basically on or off. 

[MarshallPlexi]

The output linearity is inversely related to sensitivity (how hard you have to hit the LED to make it turn on).

Because 99.44% of applications want easy turn-on and do not care about linearity, you are not likely to find anything "better".

Sounds to me like I'm going to have to really dig into vactrols then.... Sure appreciate the reply! 

[marksmartus]

Hi all. What a coincidence...I just joined the web site to talk about the H11F3N. I'm building a hexaphonic distortion box for my guitar with a Roland GK-2a pickup. I'm using the Rocket Fuzz because I work at the University of Illinois, and the IEEE at one point had a Rocket stompbox building workshop. I got all the leftover parts and boards when we moved into a new building. So it's the Rocket, but as modified by Dan Klingler for IEEE. Over the weekend I tried controlling one board with 3 H11F3N's (for Drive, Tone, and Volume). I want to be able to turn one knob to adjust the parameter on all 6 Rockets at the same time. I managed to get something that works really well with one log pot controlling the current that goes into the H11F3N's LED. It's a 50K audio taper pot in series with a 220 ohm resistor so I don't blow the LED up when I turn the pot all the way down. I will post some schematics of the wiring. The voltage dividers on the Drive and Volume settings can be replaced by a fixed resistor plus the H11F like in the data sheet page 6 where it says "as a variable resistor", "Low Frequency":

http://www.mouser.com/ds/2/149/h11f1m-185284.pdf

If you have a 220 ohm resistor in series with a 50K audio pot controlling the current into the LED, the current ranges between about 27mA and 0.14 mA. This makes the FET's output resistance measure from about 52 ohms to about 14K. You can measure it either direction with an ohmmeter and it's the same. Way cool!

For the 2 voltage dividers on Drive and Volume, I used a 10K fixed resistor and the H11F3N as described above. For the tone pot, I just stuck the H11F3N between the pins where the pot normally goes, and it worked great.

One problem is that it is current dependent. I want to have it be voltage-controlled so the behavior will be the same regardless of the number of Rocket boards that are hooked up. So I wired a pot to a voltage follower op-amp. Unfortunately, now the drive control jumps to maximum distortion very close to the end of the pot. This is because of the audio taper on the 50K pot. Changing the pot control from current to voltage control makes the audio taper in the wrong direction. I tried a linear pot, and that was better. The next thing is to try a reverse audio taper pot, which is probably the best option. I'll keep you posted.

Anyway, this H11FXN device seems to be one of the coolest things available for synth and guitar circuits now!

Mark Smart
http://www.marksmart.net

[amptramp]

Some people use separate LED emitters and photoresistor / photodiode / phototransistor / photoFET sensors just to ensure that either the emitter or the sensor can be updated more easily in the event something goes extinct.  Stick them facing each other in a tube (which could be a straw) or a piece of heatshrink tubing.  You may end up with more choices that way.

[PRR]

> current dependent. I want to have it be voltage-controlled

Voltage Controlled Current Source.

[bcnx]

Hi,

found this thread while doing reserach on the H11F1.
I need to cover a resistance range from let's say 100 ohms to 500K. This is to replace a 500K pot.
Would this be feasible?

Cheers,

BC

[R.G.]

Using a pulse width modulated scheme is much more predictable. It costs you some control logic, but you get a fully linear resistor, and change in resistance with pulse rate. In fact, you can fake a resistor by switching a capacitor. That's the way higher resistances are done in ICs.

[bcnx]

hi,

that is an intriguing idea.
how would that be done in practice?

cheers,

bc

[bcnx]

I just found https://www.edn.com/design/analog/4368893/Use-a-photoelectric-FET-optocoupler-as-a-linear-voltage-controlled-potentiometer
It seems these devices cannot provide linearity. Quite some magic is required to get them acting in a liniair way,

BC

[PRR]

> I need to cover a resistance range from let's say 100 ohms to 500K.

Did you read the datasheet?

off state resistance 300 Meg Ohms
ON state resistance 200 Ohms

Linearity of control is shown in Figure 1, and is not bad, once you sort-out the typos in the current scale.

Linearity of analog signal is shown in Figure 2. Anything over 30mV is real bent. This makes it not-usable for most audio.

[bcnx]

Hi,

I did read these maximum values in the datasheet, but these do not indicate that this component is able to emulate a potmeter with the values I specified. Especially the linearity is a question mark.
I did also see figure one, but it shows no axis value. Is this in ohms, kilo ohms or mega ohms?


cheers,

BC

[Danich_ivanov]

I've been messing around with phototransistor (not fet, buf bjt), and i was able to make it reasonably smooth by finding a sweet spot for led, it worked quite well in a vibe context. Don't know what max/min resistance was exactly, but when i tried it in a tremolo, in which it grounds the signal, there wasn't much loss in volume, which means that max was at least 50-100k. The trick is to find the right max brightness for a led, meaning not too bright, perhaps also try different led colors, otherwise action might be too steep.

[PRR]

> I did read these maximum values in the datasheet, but these do not indicate that this component is able to emulate a potmeter with the values I specified. Especially the linearity is a question mark.
I did also see figure one, but it shows no axis value. Is this in ohms, kilo ohms or mega ohms?

"Normalized to If=16mA"

The data says H11F1M max R @ 16mA is 200 Ohms.

This is a 2-terminal device. A potentiometer is a 3-terminal device, and often used 3-terminals. Clearly there is no all-purpose way to use the H11F1M as a "pot". Two might do, but need different drive signals, and probably not exactly complementary, unless you did not care about the "pot's" total value.

For nearly all audio signals, the linearity of a JFET in resistor mode is just awful.

There is no magic replacement for a mechanical potentiometer. This is why Neil Young devised his "Whizzer": servo-motors turn the knobs on his amplifier.

[bcnx]

Hi,

the circuit I have uses pots with just 2 terminals connected (Atari Punk Console), see these schematics:



This is not strictly an audio application, as we are controlling the output of a 555 IC.
I will start testing shortly and we will see of the H11F11 works out. I currently have no other ideas how to tackle my main problem, which relates to using a baby sequencer to control more than one Atari Punk Console.

cheers,

BC

[bcnx]

By he way, I beg to differ, there are alternatives to mechanical potentiometers:

https://www.sparkfun.com/products/10613

granted, it has these disadvantages:

Require a digital value, or up-down signals (cannot be controlled by a voltage)

Require a power supply, and use power

Not isolated from the control input

Limited number of discrete steps (unlike a real pot, which is continuous)

Not available at low resistance (1 kΩ minimum)

Low power (typically 1 mA max)

No negative voltages

BC