Significant Volume Drop After Adding MiniDSP Flex to System - Need Advice

[longts]

Hi Everyone,

I'm experiencing a significant volume drop after integrating a MiniDSP Flex into my system and would appreciate your insights. The system has become more detailed, but I've lost substantial output level compared to my previous DAC.

Current System Configuration​

Signal Chain:

Streamer: WiiM Pro Plus → MiniDSP Flex (coaxial connection)

• DAC/DSP: MiniDSP Flex RCA version with Dirac Live
• Preamp: Topping Pre90
• Amplifiers: 2× Topping LA90 Discrete (mono mode, XLR connection)
• Speakers: KEF R3s + SVS PB1000 subwoofer

Detailed Settings:

• WiiM Pro Plus: Fixed volume 100%, no EQ, line output 2V RMS
• MiniDSP Flex:
  • Outputs 1&2 → Pre90 via RCA-to-RCA cables (or RCA-to-XLR cables)
  • Output 3 → SVS PB1000 via RCA
  • HPF/LPF both at 80Hz, 24dB/octave
• Pre90: Set to 0.00dB (unity gain)
• LA90D: Low gain mode, 180W/8Ω per channel
• SVS PB1000: HPF off (max), Phase 0°, Volume ~10:30

The Problem​

Previous system (SMSL D6S DAC):

• Pre90 volume never exceeded -25dB for comfortable listening (when D6S volume was set at 96/100)
• D6S volume at 50-60/ 100 was sufficient (with Pre90 gain = 0.00dB)

Current system (MiniDSP Flex):

• Need to push MiniDSP volume to -10dB to -15dB for moderate listening levels (LA90D low gain, Pre90 = 0.00dB)
• Some albums (like Miles Davis "Kind of Blue") require -12dB to achieve 75dB SPL
• If I switch LA90D to high gain, the volume can be at -20dB to feel adequate
• Units running noticeably warmer (45°C)

Testing Performed​

1. Direct A/B comparison: Swapped MiniDSP Flex with SMSL D6S - confirmed ~30-40% volume reduction with Flex
2. Cable testing: Tried both RCA-to-RCA and RCA-to-XLR connections - no difference
3. Source testing: Same issue occurs with Tidal from laptop via USB (100% volume) or WiiM
4. Dirac Live: Volume issue persists even with Dirac disabled
5. Firmware: Updated to the latest version of Flex

Strange observation with MiniDSP Flex: When I set the Pre90 to output both XLR + RCA simultaneously, the volume increases slightly (occurs with both RCA-to-RCA and RCA-to-XLR cables). Note that I only connect outputs 1&2 of Flex to Pre90; output 3 connects directly to the subwoofer. This issue does not happen with the SMSL D6S.

Dirac Live Calibration Notes​

• Set master volume at -17dB to achieve 85dB SPL pink noise during calibration (LA90D low gain)
• UMIK-1 gain set to +17.5dB to get noise floor under -50dB
• My living room has significant 50Hz mode (+15dB boost visible in measurements)

The system sounds great with excellent subwoofer integration via the crossover, but the volume issue is concerning. I don't want to run the LA90D in high gain mode permanently.
Any insights into what might be causing this significant level difference would be greatly appreciated!

LA90D Low gain. Tidal. And you can see. Sound volume is very small. The speakers near by me.
Dave Brubeck - Strange Meadow Lark

 

[longts]

I tried to change LA90D with Fosi Mono V3 (low Gain), the volume is also reduced significantly compared with SMSL D6S.
I tried to remove Pre90, but the result is still the same.

 

[antcollinet]

First queswtion - are you comparing flex with RCA out against D6 with Xlr out.

If so there is an immediate 6dB difference

If you are also using Dirac in the flex then it is quite likely to have set an overall gain reduction to allow for filter boosts without clipping. Try first with turning dirac off.


Your description is also quite confusing - talking about different gains all over the place, and mixing up % with dB. % is also not a very useful unit - we have no idea what a percent value means in terms of voltage.

Your screenshots are almost illegible - photographs of the screen. Try taking an actual screen shot using whatever the windows mechanism is.

 

[Trdat]

You can lose up to 15db with frequency correction, depending on the max correction boost. Above I see you have performed DSP correction and I am presuming your previous DAC did not have that functionality.

 

[mdsimon2]

@antcollinet raises very good points.

IMO using high gain (+25.4 dB - mono) is more appropriate than low gain (+15.4 dB - mono) when using the Flex. You won't be able to drive the LA90D to full power with the Flex volume control set to 0 dB, and any cuts from Dirac will only make this worse.

Low Gain
LA90D output voltage at 0 dB volume = 2 x 10^(15.4/20) = 11.8 V
LA90D output power into 8 ohms at 0 dB volume = 11.8^2 / 8 = 17 W

High Gain
LA90D output voltage at 0 dB volume = 2 x 10^(25.4/20) = 37.2 V
LA90D output power into 8 ohms at 0 dB volume = 37.2^2 / 8 = 173 W

Michael

 

[DWPress]

DSP is a wonderful tool but the hit we take in headroom is why we need these 100 watt+ amps to keep the bass/mid bass alive. Looks like you're letting Dirac try to fill a -10dB hole around 60hz - that'll take a lot of amp power and probably not sound as good as it looks. Dirac can be tricky to implement in a non destructive way.

 

[longts]

Dear All,

Thank you so much for your kind advice. Special thanks to @mdsimon2 for sharing that formula with me. Now I understand why the SMSL D6S sounds louder—its output is 5.5V, while the Flex RCA version is only 2V.
Today I switched the LA90D to high gain and ran Dirac Live again. Now I feel comfortable with the volume at -20dB.
I got this Flex RCA version from eBay at a quite good price, and I didn't know anything about the meaning of output voltage at that time. But I will always choose balanced connections for future purchases.
I would like to attach a Dirac screenshot here for your advice.
I saw that the Pre90 can provide +16dB gain. So, should I set the gain in the Pre90 to +6dB and re-run Dirac again to reduce the workload on the Flex?
I apologize if my words are hard to understand. English is not my native language, and many audio terms are quite difficult for me to describe accurately.

 

[Keith_W]

DSP is a wonderful tool but the hit we take in headroom is why we need these 100 watt+ amps to keep the bass/mid bass alive.

FYI, you don't. Having more Watts means more headroom, it does not mean that for a given input voltage, it is louder. If a 50W amp and a 1000W amp both have a gain of 21, they will both produce the exact same output voltage for a given input voltage. The only difference is that the 1000W amp will be able to supply a heck of a lot more current if the driver demands it.

 

[mdsimon2]

I saw that the Pre90 can provide +16dB gain. So, should I set the gain in the Pre90 to +6dB and re-run Dirac again to reduce the workload on the Flex?
I apologize if my words are hard to understand. English is not my native language, and many audio terms are quite difficult for me to describe accurately.

You could add gain with the Pre90, but I don't think that will be needed if using high gain on the LA90D. I guess you could use it to add gain and use low gain on the LA90D, but I am not sure that would result in better noise performance. Given the Kef's aren't very sensitive I doubt you would have audible noise issues either way.

I don't see the need to use the Pre90 in your setup. You would typically use the Pre90 as volume control in a simple two channel system to maximize SNR, but that doesn't help you as you also need volume control on the sub.

Michael

 

[DWPress]

FYI, you don't. Having more Watts means more headroom, it does not mean that for a given input voltage, it is louder.

I never said it would be louder. However, if there is not enough power headroom will certainly diminish.

 

[DVDdoug]

However, if there is not enough power headroom will certainly diminish.

If you run out of headroom you get clipping (distortion). And at that point there is zero headroom.

If you don't use the headroom you don't need it. If you use it, it's no longer headroom! It's mostly for unknown or unexpected peaks.

 

[longts]

You could add gain with the Pre90, but I don't think that will be needed if using high gain on the LA90D. I guess you could use it to add gain and use low gain on the LA90D, but I am not sure that would result in better noise performance. Given the Kef's aren't very sensitive I doubt you would have audible noise issues either way.

I don't see the need to use the Pre90 in your setup. You would typically use the Pre90 as volume control in a simple two channel system to maximize SNR, but that doesn't help you as you also need volume control on the sub.

Michael

Thanks Michael,
You read my mind! I was actually planning to add gain to the Pre90 and use low gain mode, but I wasn’t sure if it would be good or not.
I bought the Topping bundle from a user in Vietnam at half price, so I decided to keep all of them, even though I don’t really need the Pre90. But with the RCA output from the Flex, the Pre90 helps me connect to the LA90D more easily and avoids any potential issues, as Topping recommended. If I had the MiniDSP SHD, it would be much easier with its 4V output and could handle all the other components better.
I set the Pre90 at 0.00dB and use the Flex to control volume because I have the sub connected through the Flex.
I heard someone mention that using two subs can help resolve room modes. My Dirac measurements with channel Dirac 2 show very bad results, as you saw in the screenshot. Could you give me any advice on this?

 

[DWPress]

If you don't use the headroom you don't need it. If you use it, it's no longer headroom! It's mostly for unknown or unexpected peaks.

Not at all if you are using DSP wisely,