Seed audio input specs on data sheet (Rev. 7)

Electrosmith Products Daisy Seed
1

Hello there
I am designing an audio effects pedal with the Daisy Seed Rev. 7 and was studying the Seed’s data sheet and schematic. There are a few things stated there, where I was wondering, how they came about.
This is my first post here. Please let me know if I should place this post somewhere else or if I missed some rule or some information. Also, please excuse if some questions may be a bit nitpicky.

For reference, I’m looking at the data sheet v1.1.2 and the schematic of Rev. 7.

  1. The Seed data sheet states an input impedance of 30 kOhm (typical) (on page 6, next to Figure 1.2). However, I made the following estimation. The Seed schematic shows 3.6 kOhm resistors in series with the audio inputs. The PCM3060 data sheet (audio CODEC) states an input impedance of 10 kOhm (typical). This should give an input impedance of 13.6 kOhm (typical) (ignoring the 4.7 uF capacitors).
    → Question 1a: How are the 30 kOhm derived? Am I missing something?
    → Question 1b: What is the purpose of the 3.6 kOhm resistors? Is it for compatibility with solder Seed revisions?

  2. The Seed data sheet states the absolute maximum rating for the audio inputs as 3.6 Vpp. However, I made the following estimation. The PCM3060 data sheet states -0.3 V … Vcc+0.3 V as the absolute maximum for the inputs. With Vcc = 4.5 V (as stated in the Seed data sheet) and AC coupling this corresponds to 5.1 Vpp. If we to consider the voltage divider formed by the 3.6 kOhm resistor and the PCM3060s input impedance, the limit would even be higher.
    → Question 2: How are the 3.6 Vpp derived? Is it based on the 0 dBFS.

  3. Page 5 of the data sheet shows plots for the audio performance of the Seed. I is not clear to me, what these plots are actually showing. Is it the input stage: Is it the input side (analog to digital), or the output (digital to analog), or a combination (meaning the Seed in “digital bypass”). Interestingly, the plot in the current data sheet (v1.1.2, with PCM3060) looks identical to the plots in earlier data sheets (e.g. v1.0.5, with WM8731).
    → Question 3a: What is the measurement setup for these audio performance plots?
    → Question 3b: Is it possible, that the audio performance plots are still from an older Seed revision?

2 Likes
2

I am curious about this as well, maybe someone from the team can answer?

1 Like
3

Still wondering about this. A shame that there is no answer. For anyone who wants to build hardware that uses the daisy seed these are not luxury questions unless building toy examples.

Now one could reverse engineer and experimentally find some of the answers, but that sounds not like a thing one should have to do with a project that started out as an “Arduino but for Audio”.

1 Like
4

→ Question 1a: How are the 30 kOhm derived? Am I missing something?

This seems like it is based on the previous revision (rev5) that used the WM8731 codec, which has a typical line input impedance of 30k ohms.

We’ll get this updated to match the PCM3060 datasheet + series resistor to prevent any further confusion.

→ Question 1b: What is the purpose of the 3.6 kOhm resistors? Is it for compatibility with solder Seed revisions?

This is for better gain matching with the previous revisions of the daisy seed. The datasheet describes using a series resistor to adjust input range on page 41 in the “VinL, VinR Pins” section

→ Question 2: How are the 3.6 Vpp derived? Is it based on the 0 dBFS.

This was left as the suggested maximum based on the based on compatibility with previous revisions. On the rev7 it is technically a little higher.

However, the Vcc (analog supply voltage) for the PCM3060 is from a 4v5 LDO on the seed. So technically, the the input range at the input pins of the PCM3060 would be -0.3V to 4.8V.

→ Question 3a: What is the measurement setup for these audio performance plots?
→ Question 3b: Is it possible, that the audio performance plots are still from an older Seed revision?

I’ll have to talk to the rest of the team, and see if we can dig out the original source for these graphs, and see if these were direct from the seed pins, or through one of our typ. circuits.

I believe they were (noise floor) DAC output only with zeroes being written (silence), and (1kHz Reference Sine Wave) a sine wave being generated at full-scale. So no ADC input involved in either plot.

It is possible that these are from the initial release of the seed (rev4 w/ the AK4556).

I’ll see if we can schedule some time to perform new measurements that we can publish in the datasheet (including the information about their measurement setups) to provide more clarity in the future.

1 Like
5

Thanks for the info and update! I’m no longer working on that project at the time, but an improved data sheet could be helpful for future projects (mine or other people’s).