Budget passive preamp with stepped attenuator?

[sonder]

For safety, I may add

SMD Dact type 21 Stepped attenuator 10K Passive Pre-amplifier preamp HIFI audio | eBay

Each step use 1% SMD High Quality resistors.

www.ebay.com

Or, its 50k version. But, https://www.diyaudio.com/archive/blogs/rjm/1364-rant-why-passive-preamps-totally-stupid.html sounds scary.

I should perhaps mention, I've ordered 2x different solutions already, specifically this after reading this entire thread and all it's nuances, and bit more included the specific pots within (gotcha: the new little bear mc2 is different / cheaper / different internals). I also ordered their High Precision Passive RCA only with the ALPS27 pot (the blue beauty as i understand it), and CMC816RCAs.

Both potential stop gaps, or perhaps not.

 

[solderdude]

A legitimate reason to add a stepped attenuator between DAC and amp is safety.

Yes, that is basically the only reason to do so unless one really likes the tactile feel, has small dB steps and wants this kind of device or use it combined with input switches.
Even though one can use just digital volume control (on the DAC or SW) it only takes a few seconds of 'glitching' or accidental output reset to 0dB to kill some speakers.

It also isn't a pre-amp as there is no amp, it is just an attenuator and passive one at that. Meaning cable capacitance on the input of the following amp + interlink will matter.

I have used an attenuator for years. Just a simple ALPS RK27 (10kA) and had no channel imbalance and worked flawlessly for 30 years now.
I used it to 'cap' the output to my speaker amps to get a better usable volume control range (also ALPS RK27) from an active pre-amp/control amp.
Still in use after 25 years and works flawless as well.

Best ignore Paul. He mixes porkies with truth and most people cannot tell the porkies from the truths. He is a cunning salesman.

 

[sonder]

What'll happen if I mix in a 50kA impedance pot instead of a 10k? (flow E30II in pure dac -> pot -> NC256MP)

edit: from what I can tell that's 2ohms output from DAC, 10k ohms input to amp - struggling to find any resources to calculate the optimal potentiometer k value

context:

• 10K Jenson https://www.ebay.com/itm/273591977617
• 50K ALP 09 4 channel https://www.ebay.com/itm/253646372699
• 50K ALPS 27 https://www.ebay.com/itm/253818627052

edit 2: screw it, I asked gpt4, result below:

To determine the appropriate resistance of the potentiometer that should be placed between the 2-ohm output source and the 10k-ohm input power amplifier, we need to consider the voltage divider formula:

Vout = Vin x (Rpot / (Rpot + Rin))

Where:

• Vout is the output voltage of the potentiometer
• Vin is the input voltage to the potentiometer
• Rpot is the resistance of the potentiometer
• Rin is the resistance of the load (in this case, the input impedance of the power amplifier)

We want to find a potentiometer value that will give us a suitable output voltage for the power amplifier, while also ensuring that the potentiometer and amplifier are properly matched in terms of impedance.

Assuming a standard line level signal with a peak voltage of 2V, the output impedance of the 2-ohm source is negligible compared to the load impedance of the power amplifier, so we can ignore it for this calculation.

If we want to maximize the signal level while minimizing the load on the source, we would want to choose a potentiometer resistance that is close to the load impedance of the power amplifier (10k-ohm). Therefore, a potentiometer value of 10k-ohm would be appropriate in this case.

However, if we want to minimize the noise and distortion in the system, we may want to choose a higher value potentiometer to increase the impedance seen by the source, which can reduce the loading effect and help maintain a cleaner signal. In this case, a 50k-ohm or 100k-ohm potentiometer may be more suitable, as long as the total resistance (potentiometer plus load) does not become too high and cause excessive signal attenuation.

Overall, the choice of potentiometer resistance will depend on your specific requirements for signal level, impedance matching, and noise/distortion performance.

specific calculations:

Simplifying:
Vout = 2 VRMS x (10k ohm / 10,002 ohm)
Vout = 1.9996 VRMS (approximately)

Simplifying:
Vout = 2 VRMS x (50k ohm / 50,002 ohm)
Vout = 1.9998 VRMS (approximately)

 

[somebodyelse]

Does the volume knob on AIYIMA A07 not control amplifier gain?

It does, at the expense of channel imbalance. If you lower amplifier gain the benefit of analog volume control becomes even less because you are multiplying the noise from the DAC even less

Hmm...pedantry alert...it depends what we're calling 'the amplifier.' IIRC the volume knob on the A07 is a pot attenuating the signal in front of the fixed gain buffer and fixed gain chip amp. It's not something like the baxandall active volume control where the feedback is altered. If you draw the boundary at the connectors on the back panel then it's adjusting the gain, but in the context of the preceding PS Audio comment I don't think that's quite what was meant. In the end I don't think it makes much difference though.

 

[Deleted member 58865]

Best ignore Paul. He mixes porkies with truth

Even doctors and scientists can often be found willfully lying or otherwise incorrect. The problem is that we cannot trust most people regardless of their jobs, including police officers and politicians and doctors and journalists.

I don't believe something blindly even if many media outlets repeat it over and over. Media outlets suffer the same problem of untrustworthiness. Journalists just repeat what they are paid to say. Financial/power interests can be corrupted. People's mindset around money and power is not compatible with truths. I can often hear people say they should lie that a product is good even if it is not. Otherwise, they may not sell. Even independent journalists on the internet can often be wrong.

You have to filter what's said by people. You need to have your own mental filter.

However, I try to be open-minded to new ideas I haven't heard of until I reach an understanding that they are wrong. When I hear something new, I think it may or may not be true even if it sounds far-fetched.

 

[sonder]

Even doctors and scientists can often be found willfully lying or otherwise incorrect. The problem is that we cannot trust most people regardless of their jobs including police officers and politicians.

I don't believe something blindly even if many media outlets repeat it over and over. Media outlets suffer the same problem of untrustworthiness. Journalists just repeat what they are paid to say. Financial/power interests can be corrupted. People's mindset around money and power is not compatible with truths. I can often hear people say they should lie that a product is good even if it is not. Otherwise, they may not sell. Even independent journalists on the internet can often be wrong.

You have to filter what's said by people. You need to have your own mental filter.

However, I try to be open-minded to new ideas I haven't heard of until I reach an understanding that they are wrong. When I hear something new, I think it may or may not be true.

On this, and related to my previous post, never ever trust AI chat, always verify what it says, we use it in our work and it often fabricates everything from equations through to facts and even gives legitimate looking sources which do not exist.

For example in a follow-up conversation, I asked GPT to solve more complex equations related to my audio setup, because it was correct the first time if I had naively followed it, without doing homework first, I would be putting in 63.4 volts to my amplifier. It can be very dangerous. Useful to get overviews and pointers, 100% untrustable when it comes to anything factual or important.

 

[Deleted member 58865]

https://www.diyaudio.com/archive/blogs/rjm/1364-rant-why-passive-preamps-totally-stupid.html says passive attenuator can increase system noise a lot.

Is this true or not?

 

[aikofan]

For safety, I may add

SMD Dact type 21 Stepped attenuator 10K Passive Pre-amplifier preamp HIFI audio | eBay

Each step use 1% SMD High Quality resistors.

www.ebay.com

Or, its 50k version. But, https://www.diyaudio.com/archive/blogs/rjm/1364-rant-why-passive-preamps-totally-stupid.html sounds scary.

I bought this exact product recently and am very happy with it.

 

[Deleted member 58865]

I bought this exact product recently and am very happy with it.

Good.

 

[Deleted member 58865]

To put a stepped attenuator, I need 2 pairs of shorter RCA cables. What kinds of RCA cables should I buy?

 

[sonder]

anything copper short and tight fitting, probably gold on the outside to prevent it oxidizing, relevant recent discussion here - lots of people happy with amazon basics / kabeldirekt, things like that.

before you jump at that though, you may want to read bruno's notes on noise, preamplification, rca vs xlr (snippet: The RCA connector and all it stands for should be banned by law).

 

[Deleted member 58865]

before you jump at that though, you may want to read bruno's notes on noise, preamplification, rca vs xlr.

What's the gist?

 

[sonder]

If I give you the gist, I may miss a detail that's important to your current or future setups. You can pair cheap devices to create ridiculously good audio experiences, the cost balances out however with time invested to make informed decisions. Every bit of kit has to work nicely together, and every combination is different. You pay somebody else to make the decisions and pick combinations which appeal to the lowest common denominator, or you be very specific in requirements to get the perfect setup for you. For example I've essentially built the bits I need from a ~$3000 integrated amp like a nad c399 (roon, dac, attenuator, amp) or a buchardt I150 for a small % of the cost and achieved a much better setup for me with calculable better sinad end to end, and enjoying the experience. Throw in one more variable like needing a streaming source not available via roon, or a phono input, some bluetooth, anything and it changes.

n.b. when I count hours not worked whilst researching, and factor that in rather than just "time", it works out considerably more expensive, I could have just thrown 10k at it.

 

[Deleted member 58865]

Maybe, I will read it. Still, I want to know what you are trying to say through the pdf file. What do you want to say now?

I don't care about details if I don't see a big picture.

 

[sonder]

Maybe, I will read it. Still, I want to know what you are trying to say through the pdf file. What do you want to say now?

I don't care about details if I don't see a big picture.

It arguably is the big picture.

 

[Deleted member 58865]

It arguably is the big picture.

Anyway, do you think something can go wrong by adding a 10k-ohm stepped attenuator between DAC and the amplifier?

 

[sonder]

Anyway, do you think something can go wrong by adding a 10k-ohm stepped attenuator between DAC and the amplifier?

No, I've ordered three different ones, a 256 step ladder relayed stepped attenuator (10k), a switching Alps potentiometer based one (50k), and an Alps rk27 based RCA one (10k), to compare - actually two of one by mistake.

Lots of maths, lots of reading, comfortable trying them all and will keep 2/3 of 4, eBay or sell the other(s).

I'm most confident in the stepped ladder relay attenuator, but we'll see, it should be something akin to benchmarks la4, just not 3000 dollars.

 

[Blumlein 88]

No, I've ordered three different ones, a 256 step ladder relayed stepped attenuator (10k), a switching Alps potentiometer based one (50k), and an Alps rk27 based RCA one (10k), to compare - actually two of one by mistake.

Lots of maths, lots of reading, comfortable trying them all and will keep 2/3 of 4, eBay or sell the other(s).

I'm most confident in the stepped ladder relay attenuator, but we'll see, it should be something akin to benchmarks la4, just not 3000 dollars.

I'm curious about the 256 step one. For volume control you just don't need that many steps. What is a switching Alps pot?

 

[Blumlein 88]

Anyway, do you think something can go wrong by adding a 10k-ohm stepped attenuator between DAC and the amplifier?

Devil is in the details. Probably not for most cases, but it can go wrong.

 

[sonder]

I'm curious about the 256 step one. For volume control you just don't need that many steps. What is a switching Alps pot?

The 256 step one is this, which is based on this board, it took me a couple of days to find one that wasn't from ali and looked like it had good end to end quality, n.b. it can do rca->rca, xlr->rca, xlr->xlr but not rca->xlr, which is what I'm looking for here - as if it could, it wouldn't have 2 signal paths, as you say the devil is in the detail, and thankfully both signal paths are very clear from the design. Note, historic thread on this device and topic, I asked for an update 3 years on and the original poster highly recommended it, as did other on different sites, though it's hard to find, especially w/ a good build quality.

The switched one, is the predecessor to the little bear mc2, the NS-05P (50k) which audiocircle has extensive feedback on and rollo seemed to highly recommend, it can do rca->xlr so must only have 1 signal path running through a single alps pot, so I bought 2 of them.

update: NS-05P delivered, here's the internals, overall nice weight / size / well screwed together / functions. Looks like some kind of ALPS 09?

The final one, is the 10k RK27 version of this (50k), with CMC RCA connectors, worth a punt for pure RCA->RCA.

The other two which caught my attention, but I haven't ordered yet, were the jensen relayed one (design from b whitlock paper, 10k), and another little one I won't link to, as the makers claim that three devices are fully balanced just because they have xlr on, but single pots or switchy rca in xlr out, so I don't believe it.

All in, I'm surprised this was such a hard subject to address, given there are lots of people w/ 2-4v DACs and power amps that only need 1.5vrms to blow the back off their speakers, and head.