Forum Rule: Always post complete source code & details to reproduce any issue!
Page 3 of 3 FirstFirst 1 2 3
Results 51 to 69 of 69

Thread: XLR Input on Audio Shield

  1. #51
    The shared gain is a great idea. Us the other half of 1k dual pot with the op amp by connecting it from the negative input to ground via another 47 ohms resistor and a largish capacitor (470 uF at least). The feedback resistor sets the maximum gain. Note that you should really use a "C" taper pot here, though a linear "B" works well enough. You could also reverse wire a log taper "A" pot and mark it "Attenuation" but it might be confusing and odd for the users.

  2. #52
    Senior Member
    Join Date
    Aug 2016
    Location
    Australia
    Posts
    161
    The unity DC gain idea is a good one on the op amp.

    I've been batting around which taper would be better. THAT suggests reverse audio (log) tapers for the pots, as you note, so I'll start with C (US) taper and see where that takes me.

    Because we're multiplying the gains of the two stages, B starts to look promising, as the "squared" curve somewhat approaches inverse log.

    I'm going to mock things up and see which taper, and gain setting configuration on the final op amp, provides the most "natural feeling" result. While I can model the outcome to my heart's content, the real test will be how "natural" the gain changes feel when turning the knob!

  3. #53
    Just ordered my Teensy 4.1 and audio board 😬 Super excited for this! Shout out to Paul for making these awesome products. 8 MB of flash memory 😮

  4. #54
    Senior Member
    Join Date
    Aug 2016
    Location
    Australia
    Posts
    161
    Good to hear that your T4.1 is on the way!

    I've been trying to sort out the gain issue that mhelin identified in my spare moments this week.

    So far I haven't found an elegant solution, as the THAT chip really needs a 5k reverse audio (C) taper pot. These are very rare in dual-ganged variants - none of Mouser, digikey and Element14 stock suitable pots (even in larger physical packages).

    Anything else (50k is quite common, or A / B tapers) gives a very sharp taper response (see red curve = 50k; orange is the kind of taper we want).

    As the final stage gain and this one are controlled by the same pot, the effect is squared - giving an unusably sharp taper (like the grey curve, which has 95% of the gain coming in the last 10% of rotation.)

    Click image for larger version. 

Name:	curve 9999.png 
Views:	4 
Size:	19.6 KB 
ID:	20113

    I'll keep playing with it to see if there's different solution, and also hunting for suitable pots.

  5. #55
    Senior Member
    Join Date
    Aug 2016
    Location
    Australia
    Posts
    161
    Sleeping on it provided the answer:

    - Go back to a simple switched gain arrangement for the THAT chip 20/40dB.

    - The pot controls the gain (0-20dB) of the final stage.

    Let's try and finalise things this week - so that you can get your preamp boards ordered!

    BTW: I've put the files on my github: https://github.com/palmerr23/audio_preamp

  6. #56
    Iím glad the answer came to you over a good nightís rest. By the way Iím definitely in no rush so definitely take your time.

    The files and board look awesome on github 👍

  7. #57
    By the way, does this circuit have to run off the 5V pin? Just asking out of curiosity. Iím wondering because when I run the Teensy off a battery, Iím concerned it wonít be able to supply the 5V without a boost converter if I am running off a 3.3V battery for example.

  8. #58
    Senior Member
    Join Date
    Aug 2016
    Location
    Australia
    Posts
    161
    There's no real reason why it won't work, but the current draw from 3.3v would be nearly twice that from 5V.

    I'll try the PSU modules out on the lower supply voltage and see!

  9. #59
    Senior Member
    Join Date
    Aug 2016
    Location
    Australia
    Posts
    161
    The +/- 12V boost module works OK down to 3.2V with a 100Ma load on one leg.

    At 3.0 V input the output is down to 7.5V, which is still adequate for our needs.

    I don't have one of the phantom boost modules in hand yet, but it's based on the same chip, so hopefully the characteristic will be similar.

    I hope you have a beefy LiPo battery, particularly if you are using phantom - as the current draw may be up towards 300mA @ 5V with a couple of phantom powered mics in use.

  10. #60
    Great to hear! Iím so excited 😬😬 Yes, youíre absolutely right about needing a big battery. I was planning on using 1-2 18650 cells and a TP4056 charging module. So for each cell that should hopefully be 1800 mah at a minimum.

  11. #61
    Howís it going @palmerr? Hope all is well! Just wanted to check in and see where you are at. Stay safe 🤘

  12. #62
    Senior Member
    Join Date
    Aug 2016
    Location
    Australia
    Posts
    161
    Hi Nevin,

    Doing well thanks.

    Your design is pretty stable for now, (files on my github) and I'll make any changes required if the spec evolves further.

    I'm currently distracted with an ESP32 digital power supply project, while still waiting for CS42448 audio production PCBs to arrive from the manufacturer in China.

    How are you going with the T4.1?

    Richard

  13. #63
    Hi Richard,

    Ah didnít realize haha, Iíll get to ordering parts! Sounds like a fun project you got going there. Hopefully your PCBs arrive soon!

    The 4.1 is great 👍 Teensys are super fun to play around with. Would you happen to know how to make the output of the audio board balanced by the way? Iím planning on adding two quarter inch jacks to the design to have a left and right output for my monitor speakers. You said it was relatively simple right, just flip one side of the signal I believe?

    By the way: Iím planning on ordering a few PCBs online. Iíll send a couple your way once I get them!

  14. #64
    Senior Member
    Join Date
    Aug 2016
    Location
    Australia
    Posts
    161
    Sure, any audio op-amp could be used. Here is a common configuration. It's relatively simple in concept, but needs a reasonable number of components to ensure the circuit is stable and add some low-pass filtering (usually 2-poles).

    https://electronics.stackexchange.co...nal-conversion

    There are specific chips for this, such as the THAT 1606, that allow much lower parts-count solutions.

    https://cdn.sparkfun.com/datasheets/..._Datasheet.pdf

    However, it would be very rare for any balanced input, particularly at line level, not to accept unbalanced inputs.

    - Signal to Tip.
    - Ground to Ring and Sleeve.

  15. #65
    Good to know, did not realize you could plug in unbalanced signals to balanced inputs and expect it to work. Iím just hoping there wonít be a lot of noise, but I guess Iíll see once I try it out.

    Do you have any recommendations on any good 1/4Ē jacks by the way? I think it might be cool to add a pair to the design that connects to the line out left/right at least for my purposes So I can connect to my monitors.

  16. #66
    Senior Member
    Join Date
    Aug 2016
    Location
    Australia
    Posts
    161
    I'd be inclined to do the output connections separately, as on a single board layout, you are constrained to "in one line" layouts and the board would get quite long.

    There's really no need for a PCB, just mount them as panel jacks and wire them up.

    As for sockets, I quite like the Neutrik NMJ6H series and cheaper copies. They are quite robust and the switching is very positive in action (good strong springs).

  17. #67
    Good idea, youíre right. No need for them to be mounted on the PCB. Thanks for the suggestion on the jacks.

    Iíll go ahead and convert the .brd to a Gerber file and order some blank PCBs if everything is checking out.

    Hope all is well, stay safe!

  18. #68
    Hi Richard ,

    I've put together a bill of materials, just wanted to confirm with you if you have the time before going out and sourcing parts. Here is a google spreadsheet of all the parts: https://docs.google.com/spreadsheets...it?usp=sharing

    By the way, what type of diodes should I be looking for of the 8 required diodes (just any surface mount diode)?

    Also anything else I should be looking out for when ordering these partsó such as certain types of parts to get?

    Just waiting on the PCBs in the mail now!

    Much appreciated,
    Nevin
    Last edited by Nebula525; 06-07-2020 at 02:55 AM.

  19. #69
    Senior Member
    Join Date
    Aug 2016
    Location
    Australia
    Posts
    161
    Hi Nevin,

    A quick read of the BOM looks OK, but here's one directly derived from the schematic for comparison. (It's from the standard BOM ULP in Eagle). BOM v4.xls

    The diode types are outlined on the schematic:

    Code:
    D1-4: BAV170VL very low reverse current twin diodes. 
    may be substituted by similar components (Ir <= 100nA,  Vr > 50V), 
    or with BAT54S twin (half bridge) Schottky diodes 
    (better protection, slightly higher noise).
    There are several other notes on the schematic regarding the switches, pots, etc

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •