Moderators: grovkillen, Stuntteam, TD-er
-
M*I*B
- Normal user
- Posts: 379
- Joined: 22 Jan 2018, 15:47
- Location: Germany
-
Contact:
#1
Post
by M*I*B » 23 Nov 2019, 17:11
Hi friends of short circuit,
for a new "try and error" project I have to less I²C adresses to choise. So I need a Chip to expand the single ESP-I²C to more, i.e. 4 more Ports... I know that this type of chip exist but I can't find them..
Any helpfull out there?
![Erg Blij :D](./images/smilies/icon_e_biggrin.gif)
DLzG
Micha
-
TD-er
- Core team member
- Posts: 8846
- Joined: 01 Sep 2017, 22:13
- Location: the Netherlands
-
Contact:
#2
Post
by TD-er » 23 Nov 2019, 17:31
See plugin 9 - MCP23017
plugin 19 - PCF8574
Or isn't that what you meant?
-
M*I*B
- Normal user
- Posts: 379
- Joined: 22 Jan 2018, 15:47
- Location: Germany
-
Contact:
#3
Post
by M*I*B » 23 Nov 2019, 18:09
... nope, that's wrong. I don't need more simple IO's, I need more I²C busses due I have many sensors with the same and unchangeable I²C- Adress ... The Chip simple shift the Baseadress of the Busses if I remember right...
Chip have i.e. 0xAB00 then ...
... ext. bus 1 is ACxx where xx are the adresses of the slaves at this bus
... ext. bus 2 is ADxx where xx are the adresses of the slaves at this bus
... ext. bus 3 is AExx where xx are the adresses of the slaves at this bus
... ext. bus 4 is AFxx where xx are the adresses of the slaves at this bus
-
Attachments
-
![Zwischenablage01.jpg](./download/file.php?id=4219&sid=f0f5f0aa01f17ab48df7705e170c5bdd)
- Zwischenablage01.jpg (51.29 KiB) Viewed 19537 times
DLzG
Micha
-
TD-er
- Core team member
- Posts: 8846
- Joined: 01 Sep 2017, 22:13
- Location: the Netherlands
-
Contact:
#4
Post
by TD-er » 23 Nov 2019, 19:20
Well as far as I know this is not possible.
If there would be such a chip, then we must also add support for it to make sure commands for each are sent to the right bus.
-
M*I*B
- Normal user
- Posts: 379
- Joined: 22 Jan 2018, 15:47
- Location: Germany
-
Contact:
#5
Post
by M*I*B » 23 Nov 2019, 20:10
... hmmm, I think no... If I remember right the Data are transparent. The only thing to do, in my opinion, would be to expand the bus scan area to include the extended area. For the ESP, the chip is in fact not visible. The ESP sees only the I²C clients, which only now have a different address than native
DLzG
Micha
-
M*I*B
- Normal user
- Posts: 379
- Joined: 22 Jan 2018, 15:47
- Location: Germany
-
Contact:
#8
Post
by M*I*B » 24 Nov 2019, 12:14
@ThomasB:
Yes, you hit the 1st target; thx a lot ;o) This chip does almost what I need. The good news about this chip is that you are calling all clients with the same address by choosing the correct extension port. The problem I see is that you need at least two additional ports to switch between two extension lines or, if you place a bin2hex chip in between, three extensions (see table on side 11).
I'm hopefull to find the one I described before. But if not I will give them a try .
@Flying Domotic:
Ty also for that. Looks also like a good idea, but for my projekt a too much space grabbing version...
DLzG
Micha
-
TD-er
- Core team member
- Posts: 8846
- Joined: 01 Sep 2017, 22:13
- Location: the Netherlands
-
Contact:
#9
Post
by TD-er » 24 Nov 2019, 13:11
It still needs an extension to the software to trigger the right GPIO pin for enabling the correct I2C bus.
I think it should be done by making a wrapper to the Wire interface.
Meaning it is not a plugin, but something at the core of ESPEasy, or else you have to adjust every single plugin to adapt to this.
-
ThomasB
- Normal user
- Posts: 1065
- Joined: 17 Jun 2018, 20:41
- Location: USA
#10
Post
by ThomasB » 24 Nov 2019, 19:19
The problem I see is that you need at least two additional ports to switch between two extension lines or, if you place a bin2hex chip in between
To save ESP GPIO pins, your i2c extension selection could be handled by a i2c GPIO expander chip. There are plugins for the MCP23017 and PCF8574.
It still needs an extension to the software to trigger the right GPIO pin for enabling the correct I2C bus.
I think it should be done by making a wrapper to the Wire interface.
A wrapper would be convenient indeed. But to test out this "try and error" project there may be a temporary workaround that uses rules to control the i2C extension selection and also perform the R/W to the selected i2c device.
- Thomas
-
M*I*B
- Normal user
- Posts: 379
- Joined: 22 Jan 2018, 15:47
- Location: Germany
-
Contact:
#11
Post
by M*I*B » 25 Nov 2019, 21:47
... japp ... For that T&E I will use the PCF8574 to control the PCA9518. Maybe later then I find the special chip I'm talking about, if the T&E goes to success ;o)
DLzG
Micha
-
TD-er
- Core team member
- Posts: 8846
- Joined: 01 Sep 2017, 22:13
- Location: the Netherlands
-
Contact:
#13
Post
by TD-er » 25 Nov 2019, 23:12
Hmm those are indeed brilliant
![Lach :)](./images/smilies/icon_e_smile.gif)
Only thing that may need to be changed for only some plugins is to add some extra addresses for the configuration.
-
ThomasB
- Normal user
- Posts: 1065
- Joined: 17 Jun 2018, 20:41
- Location: USA
#14
Post
by ThomasB » 25 Nov 2019, 23:28
Only thing that may need to be changed for only some plugins is to add some extra addresses for the configuration.
Oh pooh. That will indeed require a code change to a plugin with fixed i2c address choices.
But on a positive note, I just noticed that the LTC431x chips are also available in MSOP package; Substantially easier to hand solder than DFN.
- Thomas
-
TD-er
- Core team member
- Posts: 8846
- Joined: 01 Sep 2017, 22:13
- Location: the Netherlands
-
Contact:
#15
Post
by TD-er » 25 Nov 2019, 23:32
ThomasB wrote: ↑25 Nov 2019, 23:28
[...]
Oh pooh. That will indeed require a code change to a plugin with fixed i2c address choices.
Not to mention years of beta-testing...
Maybe even setup a voting committee to select the most appropriate addresses for the selection and a test panel to make sure the offered choices are political correct and will not offend anybody. (and also those without a body, just to make sure no man/woman/etc. feels excluded)
-
ThomasB
- Normal user
- Posts: 1065
- Joined: 17 Jun 2018, 20:41
- Location: USA
#16
Post
by ThomasB » 26 Nov 2019, 00:05
I found an abandoned ESPEasy pull request for adding the TCA9548A i2c multiplexer chip. It is similar to the MAX7356 chip (but easier to purchase).
https://github.com/letscontrolit/ESPEasy/pull/254
I didn't dig into the PR's details, but it is interesting; It shows that someone else has traveled on this road before.
- Thomas
-
M*I*B
- Normal user
- Posts: 379
- Joined: 22 Jan 2018, 15:47
- Location: Germany
-
Contact:
#17
Post
by M*I*B » 26 Nov 2019, 00:14
DLzG
Micha
-
krikk
- Normal user
- Posts: 118
- Joined: 28 Feb 2017, 07:57
- Location: Austria
-
Contact:
#18
Post
by krikk » 26 Nov 2019, 10:31
the TCA9548A multiplexer chip was working fine with my commit, but someone else has do update my PR to the newer codebase, sorry i do not have the time for such a project now, good luck would be really usefull if this will be integrated in the code...
Who is online
Users browsing this forum: No registered users and 0 guests