Digital rules !

[DavidDLC]

I will never never use analog potentiometers again !

Well it's better to say: I will use digital potentiometers whenever is possible, of course they are more complex and expensive to use but I'm loving them !

I mean it's not the same to be turning the volume up and down or other controls on your tv knowing there is one there compared to implement one yourself, at least that is my case.

So next step will be to replace the 10 turn pot on Dave's Electronic Load for a digital one.

Dave do you have an idea of a good number of steps for a digital pot that can suitable for your project ?

Here are some pictures:

[Zad]

Useful little things, but they can kinda fall into the "when all you have is a hammer, everything looks like a nail" category. Volume controls are my pet hate. Analogue (or at least rotary) controls are the optimal solution, but up/down buttons are cheap and easy to manufacture.

[Bored@Work]

Well, digital pots are often used as a kludge. Your usage to regulate an LCD contrast is such a kludge. A single MCU output pin with hardware or software PWM and an RC filter would do the job for a fraction of the cost, instead of that digital pot.

Same for Dave's electronic load. No need for some digital pot junk, just one output pin, a 10 or or more bit PWM, and an RC filter into the OpAmp voltage follower.

[ziq8tsi]

just wow.  i had never heard the term "digital pot", and at first i took it to mean "rotary encoder".  those are relatively expensive, but have most of the advantages of pots, plus the fact that you can turn them indefinitely in either direction.

i am totally with Zad that turning is a better user interface than clicking.  in the early 1990s it was quite common for remote controlled devices to have motor-driven analog pots so that they could be controlled locally by turning or remotely by clicking.  the move to all button interfaces quickly facilitated nasty menu-based systems like you get on modern tft monitors.

the ten turn pot in particular is a gold standard for fine user interface.  you will never reproduce the simultaneous coarse and fine control using a small number of buttons.

[JohnS_AZ]

If you look inside most modern auto and home audio equipment (with volume knobs) you'll find that they actually use shaft encoders that ultimately control digital pots or the like.

There are two big gotchas with resistive analog pots; They are mechanical assemblies and thus expensive to manufacture, and being mechanical they are subject to wearing out and getting dirty (thus injecting noise and intermittent failures into whatever circuit they are controlling). 

The gotchas with digital pots are that they are pretty limited in how much current they can handle, and by definition they do not offer an infinite range of settings.  Digital pots are great for setting volume in an audio circuit, but you'll never see them used for, say, calibration points of instrumentation amps.

Resistive pots are great. Digital pots are also great. It's all about the application.

All that being said, nothing can match the feel of a big heavy knob spinning a high quality wirewound pot. 

[mikeselectricstuff]

the ten turn pot in particular is a gold standard for fine user interface.  you will never reproduce the simultaneous coarse and fine control using a small number of buttons.

You can get pretty close, maybe better with a  rotary encoder and carefully tweaked velocity sensing.

[DavidDLC]

Well, digital pots are often used as a kludge. Your usage to regulate an LCD contrast is such a kludge. A single MCU output pin with hardware or software PWM and an RC filter would do the job for a fraction of the cost, instead of that digital pot.

Same for Dave's electronic load. No need for some digital pot junk, just one output pin, a 10 or or more bit PWM, and an RC filter into the OpAmp voltage follower.

Well I don't know if you can input a PWM to the LCD contrast input, the specs on the LCD requires a potentiometer to be connected, are you getting confused with the backlight and contrast ?

[Zero999]

The gotchas with digital pots are that they are pretty limited in how much current they can handle, and by definition they do not offer an infinite range of settings.  Digital pots are great for setting volume in an audio circuit, but you'll never see them used for, say, calibration points of instrumentation amps.

They also have poor tolerance and a very limited voltage range which has to be within the power supply rails.

[hans]

I think you're missing the point of Dave's simple design. The opamp needs a reference voltage which it wants the feedback to control to. It will regulate the transistor untill it reaches it's reference voltage. Because it's a voltage, not a resistance, you can use any DAC you want. The potmeter is used to get a voltage really easily. If you want to digitally control this thing completely (I.e. make a variable load controlled by USB) you could better get like a SPI 10-bit DAC (1024 steps) or a PIC with one.

If you'd use a more cheaper digital pot, like 256 steps, you can only control every 4mA.

Digital pots aren't that accurate really. 25% tolerance (min to max resistance!), and if you take a 10-bit version you'll pay an impressive amount of money (6 euro's here for the MAX5481).  And how much is a dedicated 10-bit DAC? Like 2,50 euro for the AD5611, which can change 'value' at 1.7M times per second.

Analog pots are just the best components to adjust and calibrate analog designs, especially when high accuracy is needed. You could use digital pots if you have a schematic that requires a real resistance instead of a voltage to control it. In case of dave's design, it requires a voltage, so anything that produces a voltage (DAC, resistive divider with pot) will work. Then just is the question, which one is cheapest/best/reliable

[DavidDLC]

I see the point for the Electronic Load, I will just use the 10 turn pot instead.

 

[Mechatrommer]

i was sick with a car radio cassette to control volume with up and down button. same with car aircond system to control temp. though classic, a rotating knob is still a winner! i can easily find and reach the knob while my eyes are on the road. with buttons, i usually hit the other wrong button.

[eternal_noob]

Analog pots are just the best components to adjust and calibrate analog designs, especially when high accuracy is needed. You could use digital pots if you have a schematic that requires a real resistance instead of a voltage to control it. In case of dave's design, it requires a voltage, so anything that produces a voltage (DAC, resistive divider with pot) will work. Then just is the question, which one is cheapest/best/reliable

The really bad thing, in my opinion, is that the components industry has never got around to producing nice geared vernier drives for potmeters in mass scale. Sometimes 10 turns or even 20 is way too little, and 100 or 200 would be more suitable.
A pot with a high def. plastic track with 100 turns and a two speed gear mechanism would be a dream... 

[alm]

The really bad thing, in my opinion, is that the components industry has never got around to producing nice geared vernier drives for potmeters in mass scale. Sometimes 10 turns or even 20 is way too little, and 100 or 200 would be more suitable.
A pot with a high def. plastic track with 100 turns and a two speed gear mechanism would be a dream... 

Don't forget that the number of turns is a trade-off between fine adjustment and speed. For example, if you had a 100 turn volume control set to the max, it would be a lot of work to turn the volume down to take a phone call.

Bob Pease claims in his book 'Troubleshooting Analog Circuits' that single turn trimmers actually actually drift less and have better resolution than multi-turn trimmers that would change value after some minor vibrations.

[eternal_noob]

Don't forget that the number of turns is a trade-off between fine adjustment and speed. For example, if you had a 100 turn volume control set to the max, it would be a lot of work to turn the volume down to take a phone call.

But.. but... this is what wifes are for, isn't it? 
I'm thinking servo controlled coarse adjustment and a hand cranked fine. Maybe like a rotary selector for each ten percent. You know; 'zip' zip' zip to 60 percent and then crank the rest or something like that.