[sdiy] nice Digi Pot

[rsdio at audiobanshee.com]

Daisy-chaining is rarely necessary. Usually, it's just a waste of a pin. If your main processor's MOSI has sufficient fan-out, you can simply connect it directly to the input of every SPI slave. If the fan-out is weak, then an external buffer will suffice.

The only time daisy-chaining is really useful is when you have a serial-to-parallel converter, and you want downstream parallel devices to have different bits out of a larger bit string. But that's more than simple daisy-chaining, it's also delaying the data by 8 bits (or whatever the parallel output width might be) within each chip.

I don't see why you would want to run multiple 8-bit DPOT chips off a single, larger bit-depth word command. It would be more trouble to combine multiple DPOT settings into a single word than to simply write each value separately - thus there's no real need for daisy-chaining.

Looks like a nice DPOT. I've been using the MAX5481/MAX5482 pot, MAX5483/MAX5484 variable resistor, and AD5290, but they're not necessarily ideal for the OP's project. The MAXIM parts are 10-bit, in case you need extra resolution beyond 8-bit, and have non-volatile memory so that they power up in the same position as last saved. The Analog Devices part has support for an external zero-crossing detection circuit to avoid zipper noise, but it's not an integrated feature.

Brian


On Jun 27, 2016, at 12:40 PM, Vladimir Pantelic <vladoman at gmail.com> wrote:
> "This SPI Interface does NOT support daisy chaining."
> 
> :(( why Microchip, oh why....
> 
> On 27.06.2016 19:06, sleepy_dog at gmx.de wrote:
>> Reading the thread "non-mechanical switching solutions",
>> I just remembered a nice series of digital potentiometers which I discovered
>> some months ago,
>> which are SDIY friendly ( for those not afraid of TSSOP ;) ), as they
>> - support +/- 18V dual rails
>> - digital voltage: 2.7 ... 5.5V (or even 1.8V)
>> - cost less than 2 bucks
>> 
>> I have used, as an experiment, digi pots in some VCFs to easily control
>> resonance from an MCU, but before I found those here, I only knew similar
>> products in a price range of 8...12 US$ or something like that.
>> 
>> So I just thought, I'll throw that in here, maybe someone else also only knew
>> the expensive ones.
>> 
>> MCP41HVx1 series, where
>> http://www.tme.eu/de/details/mcp41hv31-503e_st/digitalpotentiometer/microchip-technology/
>> 
>> 
>> available in 5k, 10k, 50k, 100k
>> 
>> datasheet
>> http://www.tme.eu/de/Document/8b361619149495099964ddab197bebfe/mcp41xvx1.pdf
>> 
>> 
>> regards,
>> Steve
>> 
>> 
>> Am 27.06.2016 um 16:27 schrieb Vladimir Pantelic:
>>> On 27.06.2016 16:15, Pete Hartman wrote:
>>>> First thought: CD4052, but that doesn't work well in practice at all.
>>>> 
>>>> Second thought: DG409, better characteristics, but that still doesn't
>>>> work sufficiently well.
>>>> 
>>>> Dig around a while for why.... aha, the on resistance for both is
>>>> significant, especially for the capacitive stages of a ladder filter.
>>>> The DG409 has on the order of 100R - 120R.  Too much, and confirmed to
>>>> be the issue by comparing physical connections with wire to physical
>>>> connections with 100R resistors.
>>> ADG1414, R_on = 9.5R and 8 switches inside one package - but needs SPI
>>> so you need to add a 50¢ uC to control it
>