Modhouse Tungsten Measurements & Official Discussion

Regarding my earlier comment with V222 distortion. I realized by accident that this seems to have had something to do with the DAC I was using. I was using the Geshelli Dayzee via XLR at the time that I reported the V222 distorting at +18 pre gain and struggling with volume on some tracks at +12. Last week, I was testing a cable quickly and threw the V222 on my Denafrips Ares 12th with the Tungsten, and it was picture perfect at +12 pre gain. I was really surprised by this. The Dayzee is a beefy DAC, but there is no listed spec for the voltage of the outputs. The Ares 12th is either 4V or 4.4V on XLR I think. In any case, the V222 + Ares 12th via XLR output works perfectly with the Tungsten.

I’ve never run into a situation like this where the DAC + amp is causing a problem, but it is possible that I have unfairly blamed the V222, where it could be some issue with the combination of the two or just the Dayzee not getting to an adequate output level, as the Ares 12th just sounds wonderful with the V222 and Tungsten.

1 Like

Update 12/14/24 (previous post was June 21, 2024): The Tungsten run just fine on my setup as described above. Plenty of power, plenty of voltage, plenty of headroom. It is a headphone that is balanced, well-tuned and has strong technical chops for the money! I have since purchased a Susvara (sold the HE 1000v2) using the same chain. To me, the Susvara is as all the reviews state - the best. From my experience, the Tungsten is a top dog from a value perspective. It does not compete with true TOTL cans which IMO, start around $3k and go up.

Wanted to add some math/info for pairing these with the Schiit Midgard in particular.

I’ll split the info into two sections:

  1. What the Tungsten needs at a certain volume
  2. What the Midgard can provide at a given impedance

Section 1:
We know from the store page that the Tungsten has a sensitivity of 78 dB/mW and 155 ohms impedance.

If we assume an average listening level of 80 dB (what I listen at) and a crest factor of 30 dB (enough for most genres of music) we want to see what the Tungsten needs to produce 110 dB.
We can get this from the calculator:

Section 2:
Disclaimer: The correct way to do this is to use an audio analyzer like the AP x555b with a 155 ohm load on the Midgard and see what we get. Below, we’re seeing what the math gets us as an estimate

The Schiit Midgard provides 3200 mW into 50 Ohms

and 750 mW into 300 ohms.

Using the values for Current and Voltage above, we can see that voltage ranges from 12.6V to 15V and Current ranges from 0.25A to 0.05A.

I’m not an electrical engineer but using Amps and Impedance to calculate Voltage and Power gives some nonsensical results (155 ohms and a current of 0.25 A gives a value of 9 watts which doesn’t make sense as that is higher than the power outputted at a lower impedance). Given that Voltage induces current and not the other way around, we’ll use the voltage values from above (12.6 V at 50 ohms to 15V at 300 ohms) to see what we get at 155 ohms.

At 12.65 V @ 155 ohms:

At 15 V @ 155 ohms:

So we can estimate that the Midgard puts out roughly 1 ~ 1.45 W into 155 Ohms. We can call this value P1

Now, as Oratory1990 explained to me (but this also seems to be standard EE knowledge), the change in sound pressure caused by changing the power from P0 to P1 is calculated as 10*log10(P1/P0) where P0 is 1mW and P1 is the power output at a given impedance.

In Oratory’s words ā€œIf you know the power sensitivity (in dB/mW) and you want to know how much sound pressure the headphone is producing at an input of X mW, you calculate the difference in level caused by changing the power from 1 mW to X mW, and then you add this difference to the sensitivity.ā€

As we calculated above, P1 in our case ranges from 1000mW to 1450 mW.

So we get the following values:
@ 1000mW: 10 * log10 (1000/1) = 29.9 dB
@ 1450 mW: 10 * log10(1450/1) = 31.6 dB

Given the Tungsten’s sensitivity of 78 dB/mW we get the following:
@ 1000mW: 78 dB/mW + 29.9 dB = 107.9 dB
@ 1450 mW: 78 dB/mW + 31.6 dB = 109.6 dB

What this tells us is that the Schiit Midgard produces between 107.9 dB to 109.6 dB on the Modhouse Tungsten at max settings. This is just shy of the 110 dB target we were aiming for.

TL;DR: If you listen at an SPL of 80 dB on average, the Schiit Midgard should be good enough to power the Tungsten for most genres of music (assuming the crest factor in your music is <= 30 dB).

4 Likes

In case you need to do this for another amp the quick way is to just determine the voltage limit (it’s very unlikely an amp will be current limited at 150 Ohms).
Just take the highest impedence rating 600 Ohms 375mW and use V^2/R = P.
V = SQRT(P*R) so SQRT(0.375 * 600) in this case 15V RMS.
You can then just type the numbers into a calculator like this one

And you’ll get 109.76dB as the maximum volume as you calculated.

2 Likes

Oh that’s so convenient haha. Does this usually align with experimental results from an audio analyzer?

Also, what is meant by ā€œvoltage limitā€? Is it the max voltage an amplifier can provide? If so, is that a fixed attribute of the hardware that needs to be measured or something that can be calculated?

Should be close enough, assuming the quoted specs are accurate.

Power is a misleading spec in amplifiers.
When you design and build an amplifier, you effectively design for a maximum voltage, based on the power supply and a maximum current, usually based on some limitation of the circuit design, where the designer likely picked an optimal load, which usually has as much to do with heat dissipation and packaging as it does the power supply.

Voltage is volume. Though note that dB is a logarithmic scale so doubling or halving the voltage only gives you a 6dB change (Which is where the whole the first W is the most important stuff comes from).

Those are the limits, of the amplifier, it can’t exceed it’s voltage limit or it’s current limit, for low impedance loads the limit will be the current that can be delivered and for high impedance loads it’ll be the voltage.

Lets take an example An amplifier uses a 20V power supply, that means it can only deliver +/- 10V peak (it’ll be bit less than this), or 7V RMS lets sat the designer wants maximum power into say a 32 Ohm load he’d limit the current to 10/32 or about 0.3125A, he’d probably round up, but it doesn’t really matter.
So into 32 Ohms he gets the expected full 7V RMS so 7*7/32 W or about 1.5W RMS of power.

Because you have to follow Ohms law at 16 Ohms you can still only deliver 0.3125A so you can only deliver a maximum 16 * 0.3125 or about 5V peak or about 3.5V RMS so about 0.75W.

Again following Ohms law if the load was 64 Ohms, that would support a voltage of 20V Peak, but your limited by the power supply to 10V (7V RMS), so you end up with 7*7/64 or about 0.75W.

You exceed the maximums you clip.

Transformer coupled Tube amps don’t exactly work like this because the load is a part of the circuit and the output is transformer loaded (coils store energy).

4 Likes

So given that the Midgard says "Power Supply: ā€œWall wartā€ style 14-16VAC transformer, stacked rails of +/-20V and +/-30V regulated voltage rails, 10,000uF total capacitance" what can we say about the maximum voltage the designers had in mind?

1 Like

Well from the power numbers we know it’s putting out 15V RMS or roughly +/- 21V before it hits whatever noise threshold they picked
It honestly could be using either the +/-30 or +/-20 at the output, though most likely the +/-30 is for the output stage and the +/-20 for it’s input stages.

All the power supply voltage tells you is what the absolute maximum could be, the circuit itself will steal some of that headroom. For example the last amplifier I designed and built has a +/- 9.5V swing from a 20V power supply before it hits about 1% distortion, the remaining 1V or so is lost to the CCS on the source of the Mosfet.
But that’s about as low a loss as you can get, some circuits can be as bad as 50% of the supply rail. There are a lot of choices made in amplifier design.

2 Likes

One more follow up question, did you choose to go with the highest impedance figure because that’s where voltage is tested the most? I.e. if voltage is electrical pressure, the highest resistance scenario is where voltage limits can be expected to be encountered?

Yes, you pretty much guarantee that the power is voltage limited at the highest quoted impedance.
But you’d get the same result if you picked any impedance above whatever the peak power value is.
Anything below that will be current limited.

I guess you can think of it that way.

Basically it’s all Ohms law V=IR and the definition of Power = IV
combining them you get Power is V^2/R or I^2R
So as you increase Resistance the amount of current required for a given Power decreases, and Voltage required increases.

4 Likes

Just got my Tungsten DS headphones and was a bit nervous about driving them. That power chart really helped me feel more comfortable. My GSX-Mini seems to do a good job at my listening level, even when using the Roon headphone EQ profile which knocks down the gain a bit.

Using Balanced in to the GSX from a Topping E50 and balanced XLR out of the amp.

3 Likes

I finally got my Tungsten DS from the May orders of last year.

I had to sign for them and get to work, hopefully listen to them tonight! :+1:



6 Likes

Very nice!

But holy cow, order in May and delivered at end of March? Yikes. I think I placed my order in August of last year.

1 Like

September for me. Guess I’ll get them just in time for my birthday. :roll_eyes:

3 Likes

IDK if either of you got his recent email updates for orders, but he said he’s getting his order backlog reduced through streamlining certain processes, and taking fewer orders during each months order window.

He’s working on June orders now, which were about half the number of May.

Hopefully yours arrive in a more timely manner than mine, but good things come to those who wait (supposedly) :wink:

1 Like

I’m not getting any email updates :frowning:

But I did know about efforts to reduce the backlog. It’s one of the reasons I couldn’t order mine in custom colors.

I’m actually pretty patient about it, although an update would be nice. I’ve heard nothing since the order confirmation.

1 Like

Yeah it seems like not everyone was getting them.

I got 1 in January and another in March, and was posting them on the tungsten thread on headfi.

If I get any more I can pm it to you if you want, just let me know.

Since I got the headphone now I might fall off the list though…

I can’t find the Tungsten thread on Head-Fi, only the review thread. Please post a link to it.

1 Like

Sure no problem. January update page 284.

1 Like

For anyone that owns or is getting a Tungsten, you’ll want to watch this vid:

3 Likes