0
\$\begingroup\$

I'm not quite sure how level and impedance bridging works so I am hoping for an illuminating answer!

I'm generating audio with a PCM5102A DAC. The datasheet of the chip specifies a 2.1V RMS output signal at full volume, and a minimum "load impedance" of 1kΩ.

My questions are:

  1. For equipment designed to take instrument level inputs (e.g guitar amps, DI's etc) what RMS voltage level and input impedance is expected? I.E how high is instrument level as opposed to line level, mic level, etc?
  2. What are my options for converting the DAC output into an instrument level input? This is for an audiophile application, so sound quality is a high priority. It's also battery powered, so power efficiency is important too. Is a high quality transformer the way to go? If so, how do I do the maths to pick the right transformer?

Thanks!

\$\endgroup\$
2
  • 1
    \$\begingroup\$ Please split these two questions into, well, two. This page about instrument vs. line levels answers your first question. It really ought to be answered somewhere on this forum, but I can't find it! \$\endgroup\$ Commented Jan 24, 2019 at 0:57
  • \$\begingroup\$ The question is about converting a DAC output to a guitar amp input, both points are relevant and should be kept together. But I agree, there should be another question on the forum that explains all the different signal levels and how they work. \$\endgroup\$ Commented Jan 24, 2019 at 1:41

2 Answers 2

Reset to default
2
\$\begingroup\$

In general, if it's an instrument output it's not hugely well controlled impedance, and it doesn't have to be matched. I wouldn't even try to impedance match: I'd just use an L pad that presents 1k\$\Omega\$ to the DAC, attenuates down to whatever I've decided to call instrument level, and has whatever (nicely low) output impedance that works out to.

Below is an example 20dB L pad (it looks like an upside-down L -- it got the name before anyone thought to call it a "Gamma pad", even though it looks like an inverted \$\Gamma\$). Allmost all you need to do is make sure the resistances add up to 1000\$\Omega\$ and have the attenuation you want.

But before you use this circuit, you should make sure that it'll work with the DC paths to ground as presented -- you may need to add blocking caps for it to work in your particular situation.

For the attenuation \$A\$ as a ratio (and if I'm getting my math right), \$R_1 = (1000\Omega)(1-\frac{1}{A})\$ and \$R_2 = (1000\Omega)\frac{1}{A}\$. Or just put a 1k\$\Omega\$ pot in there. Or use a 200\$\Omega\$ pot for R2 and an 806\$\Omega\$ resistor for R1. Or something.

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
1
  • \$\begingroup\$ Thanks Tim - could you explain that a bit more clearly for me? What is an L pad, how would I design such a thing and how would I calculate its' values? \$\endgroup\$ Commented Jan 24, 2019 at 1:37
0
\$\begingroup\$

I'm not sure whether a resistive L pad (as posted by Tim above) or a transformer is the best option to go with, so I'm posting the latter for completeness:

Transformer circuit

Assuming that a roughly "instrument level" voltage is about 0.17V RMS, to get that from 2.1V RMS requires a turn ratio of $$\frac{V_{in}}{V_{out}} = \frac{2.1}{0.17} \approx 12:1$$

The relationship between the impedances is given by (assuming I got my maths right) $$R_{out} = \frac{R_{in} V_{out}^2}{V_{in}^2}$$

So if the DAC requires a load of at least 1kΩ, the amp / DI box / whatever connected to the jack output needs an impedance of at least $$\frac{1000 \times 0.17^2}{2.1^2} = 6.5Ω$$ which will almost certainly be fine with almost any equipment?

\$\endgroup\$
3
  • 1
    \$\begingroup\$ I doubt you could get a really audiophile transformer that would handle such low-level signals. \$\endgroup\$ Commented Jan 24, 2019 at 6:05
  • \$\begingroup\$ It seems that they are available, but they're bulky and expensive. The L-Pad answer is the most practical unless you are going for a specific "colouring" of the sound \$\endgroup\$ Commented Dec 18, 2020 at 1:37
  • \$\begingroup\$ Looking at those two DC shorts that you've made, and re-looking at my answer: I think we both need to add DC blocking caps to our answers. \$\endgroup\$ Commented Dec 18, 2020 at 2:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.