Needing more than a spark test?

[graham-xrf]

Mark, the ADCs seem to be more than "looks pretty similar". They use so many identical pinouts, it may be possible let either go on the same layout, with only one pin perhaps linked to a different function if it is used at all. The basic difference is that the 7622 has a differential input, so IN- [39] becomes INGND [39], and IN+[43] becomes plain IN[43] for the AD7667 single ended input. The inputs still use the same pin numbers.

BYTESWAP is pin [4] on both ADCs, shown used in the AD7667 Fig26 Typical Connection Diagram, but not apparently invoked in the AD7622.
Pin [7] is called IMPULSE, on the AD7667, and called /NORMAL on AD7622. Both are connected to GND anyway.

I was planning to load in the samples in serial manner, because they can still load in at 1MSPS, or at 2MSPS if "WARP" mode is invoked, only because the Raspberry Pi can do it, but parallel is OK into a Teensy as well.

The group pins 5, 8, 6 connect to 0VDD in my AD7622 circuit, whereas the same pins are taken to DVDD for the AD7667.

I did mark up the Fig25 typical connection diagrams with blue pin numbers in a separate PNG image, because I found it handy instead of repeatedly looking up pins on the 48-pin LQPF picture. The corresponding connection diagram for AD7667 is on it's Fig 26.

Note that pin 3 is used for for DGND on AD7622, and is just not connected to anywhere on AD7667.

The layout, so far, seems to be a pin for pin compatible, drop-in replacement when it comes to the layout, so long as one looks after what happens to pins 5. 6, 8, done with a zero-ohms link.

I include both pictures here. The AD7667 shows the interesting pins in a different colour, and the pin function names, where there are differences in usage, are highlighted in red.

I usually end up needing two or three print-outs of these sort of pictures, because mine get pretty messed up when I try stuff out. So far, it's one little scorch mark, curled up corners, and a bit of a coffee mug ring + spillage.



Compare to the AD7622 one I used..

 

[homebrewed]

Interesting that the typical connection diagram you found for the AD7677 is a bit different than the one I have. The most significant difference is that 10uF capacitor hung on the Ref pin. Mine doesn't show that.

 

[graham-xrf]

Interesting that the typical connection diagram you found for the AD7677 is a bit different than the one I have. The most significant difference is that 10uF capacitor hung on the Ref pin. Mine doesn't show that.

I downloaded it from analog.com this very evening!
Ahh - I see it now. On your datasheet, Figure 26, that capacitor is Cr(4), indicating look at NOTE 4.
"A 10uF ceramic capacitor, (X5R, 1206 size) is recommended (eg Panasonic ECJ3YB0J106M)"

BTW - we are seriously pushing at 100 pages of this thread.

 

[homebrewed]

I downloaded it from analog.com this very evening!
Ahh - I see it now. On your datasheet, Figure 26, that capacitor is Cr(4), indicating look at NOTE 4.
"A 10uF ceramic capacitor, (X5R, 1206 size) is recommended (eg Panasonic ECJ3YB0J106M)"

BTW - we are seriously pushing at 100 pages of this thread.

Aha! I took the DS off the Digikey web site. Apparently their information can be stale. I'll head over to ADI and get the latest one to see if there's anything else I need to be aware of.

I don't think we're even close to having the longest thread, but some of the really long ones are more generic like POTD. I'm not so sure for a (mostly) single subject.

 

[homebrewed]

If any of you are seriously looking at getting a TFT display for this project make sure its logic/power supply requirements are compatible with your particular processor board. Adafruit has different versions of their 3.5" TFT display, but only some of them appears to be compatible with both 3.3 and 5V systems. They are ones designed around the ILI9341. I bought the right one for my Teensy4.0 but it would be all too easy to get the wrong one. The one offered by PJRC also is based on the ILI9341.

I just fired mine up using an old Uno and it worked fine for that. Next up: I'll try it with a Teensy LC, which is a 3.3V processor. I'm trying to avoid potentially damaging my one & only T4.0, although I do have a 4.1 on the way.

I haven't checked the PJRC graphics library but the Adafruit one allows you to rotate the axes in 90 degree increments. Handy because I want the long dimension to equal the X axis (time) in this particular application.

One important thing to note is that, like many video display drivers, the (0,0) point is the upper-left corner on the display. So for plotting data in a landscape display mode the Y axis has to be inverted: Y' = (239 - Y). The Y axis pixel "coordinate" range is 0....239, that adds up to 240 pixels

 

[WobblyHand]

The PJRC ILI9341 is compatible with Teensy 4.x and 3.x. On the anti static bag is pasted a sticker which tells you that the screen pixel map is 180 degrees out from the display. Both are simply rotated with simple commands. In the beginning it is a bit fun to sort that out, but I can assure you it can be done successfully! I power the screen with the 5V input, but the screen accepts 3.3V I/O. Internally, the screen runs on 3.3V.
If I recall the screen rotation is just something simple like tft.rotate(n); where n is an integer, 0, 1, 2, 3. I think the touch display is similar.

I was at a Microcenter store this morning. They had Teensy 4.0 and 4.1, and not a single Raspberry Pi of any sort. Plenty of accessories for Pi's but no Pi's, which is what I expected to see. The Teensy's were reasonably priced, primarily because PJRC has a similar policy as The RPi Foundation, no scalping. There were no Teensy 3.X that I could find.

 

[homebrewed]

The PJRC ILI9341 is compatible with Teensy 4.x and 3.x. On the anti static bag is pasted a sticker which tells you that the screen pixel map is 180 degrees out from the display. Both are simply rotated with simple commands. In the beginning it is a bit fun to sort that out, but I can assure you it can be done successfully! I power the screen with the 5V input, but the screen accepts 3.3V I/O. Internally, the screen runs on 3.3V.
If I recall the screen rotation is just something simple like tft.rotate(n); where n is an integer, 0, 1, 2, 3. I think the touch display is similar.

I was at a Microcenter store this morning. They had Teensy 4.0 and 4.1, and not a single Raspberry Pi of any sort. Plenty of accessories for Pi's but no Pi's, which is what I expected to see. The Teensy's were reasonably priced, primarily because PJRC has a similar policy as The RPi Foundation, no scalping. There were no Teensy 3.X that I could find.

The Adafruit graphics library has an identically named command. To test the pixel vs. coordinate assignment I just issued tft.rotate(n) plus a tft.write() command to display text correctly, then drew a line from (0,0) to (319,239). Observation did the rest.

 

[homebrewed]

Unfortunately the Adafruit display draws more power than the LC's 3.3V regulator can supply. The T4.X boards are OK though.

 

[WobblyHand]

For my ELS, I powered the display with 5V, not 3.3. That way the onboard regulator on the processor is not used to power the display. Basically USB powered it, not the Teensy.

 

[WobblyHand]

The Adafruit break out display can be powered by either 3.3V or 5V. My chronometer uses the 5V option, so I don't draw current on a Teensy regulator. Should work the same, I think we have the same displays.