Lii Audio 15" full range

The miniDSP makes my system sound like a mass-market receiver from the 80s... I cant run the high pass through it - wrecks whatever magic was happening between the Liis and the Jolida LM1875 amp.

I have a sneaking suspicion you might benefit heavily from using a good buffer.
Getting something as simple as a couple CMOY boards from ebay might do the trick.
The output stages in the basic minidsp units sound good enough with high input z amps. I suspect your LM1875 has a bit low input z, probably makes the minidsp sound a bit anemic.

Remember to set all outputs correctly, easy to mix it up if you're a bit quick.
 
I have a sneaking suspicion you might benefit heavily from using a good buffer.
Getting something as simple as a couple CMOY boards from ebay might do the trick.
The output stages in the basic minidsp units sound good enough with high input z amps. I suspect your LM1875 has a bit low input z, probably makes the minidsp sound a bit anemic.

i was gonna say that it sounsds like an impedence mismatch and a preamp might help. the minidsp's as DAC are not the best at all, but they are not too awful.

i had their top of the line SHD and while I eventually diceded to move something more expensive, it was not a bad unit. if you use it for digital crossover and room correction and not for D/A conversion then they are very good.
 
@Kaffiman - The Jolida amp is basically an LM1875 chip amp with a tube stage in front of it. There's a high voltage coupling cap between the tube stage and the LM1875. I imagine the two triodes per channel are arranged in a totem-pole topology, so I imagine the input impedance is whatever the volume control is. I've never drew out the schematic for it.

The LM1875 part also has a dc zeroing design that uses a cap to ground in the feedback voltage divider. Currently I have that shorted. These two highpass filter positions could be exploited as part of an active crossover.

I got the 4, 1.0 uf film capacitors and used two on the input and two on the output. I modeled the 6k input impedance the miniDSP supposedly presents and the rolloff isnt too bad considering I have no hope of making it below 30 with the OBs. Sound is a step up, but there's still a definite processing effect on the sound, that blurs something - compared to a cable. So test # one (double d/a) is FAIL.

It's listenable, it's just that If I can hear it at 63...I wont put something in my signal chain that would hold back these F15s from what they can do.

@Dadbeh - "minidsp's as DAC are not the best at all, but they are not too awful". It's the kind of thing that if I didnt hear it first, I could easily plod along thinking "It sounds pretty good!" and be listening like that for years. But if compromised by the actual DAC chip, they cut the odds back substantially that a I2S connection will help its sound.

I'd still like to try. Does anyone know of a straightforward way to connect USB (appearing as a sound card to the pc) and provide I2S output to the miniDSP? Anyone actually do this? Any help would be appreciated. Thanks!
 
Last edited:
IMHO you're trying to make a silk purse from a pigs ear ; the MiniDSP is alway's going to limit you. Just use it 'out of the box' to empirically work out your xover, equalisation, gain etc. then build the pass jfet crossover, from the DIY Audio store, with those same parameters.
 
@nautibuoy - "the MiniDSP is alway's going to limit you" - OK, #$%^, so there's a $100 lesson from the "buy before actually hearing it" dept - Hope the next thing I've ordered that way does a little bit better for me -

When I set it up for 24db LR, 100Hz sounds like a good number. Got some work to do to put it back to factory; watch for it on ebay...
 
Hi Joe,

Are you attempting to use a full range unit above an open baffled bass driver?

If so, are you aware that the crossover frequencies for each driver are (normally) different?

I've attached the MJ King spreadsheet of a design that I've built. It used minidsp, and I was happy with it. You can see that the LF cut off is 110Hz, and the HF cut off is 285Hz. Note also that the order of the crossovers is different. As you can see, the predicted frequency response of the system (on the penultimate page) is reasonably flat. I confirmed this with REW measurements.

I'm sorry if none of this is appropriate or if you already knew the above!

Regards,
Ian
 

Attachments

  • U_Frame_Corner_Active QTX 902-539 and TC9FD18-08.pdf
    239 KB · Views: 83
Well, considering costs of all the analog LR crossovers I've looked into (including suggested) I might as well just keep the miniDSP for LF duties.

Even if I went ahead with the analog, I'd still have to figure out how to do the reverse rising response in addition to the low-pass cutoff - and implement it in whatever circuit without disrupting the LR filter. That part's done already in the miniDSP -

So I put a post in the miniDSP forum, asking what can I do. (Buy a HD...)

Meanwhile, I'll suffer the atrocity of simply running the F15 full range and use the mini to tune the assist woofer's response around its natural rolloff on the OB.

I wont hear the fuzzy / squished soundstage doing that...
 
I forgot to mention that - of course - I went to an analog wall-wart power supply, pulling the PC USB cable out while listening.

@Ian - Yep, I'm trying to do the same as you've done. Appreciate the info. Because I dont like the sound of the mini as a high-pass filter, I'm stuck with either running the F15 full range, customizing the amp, or getting a two channel active high-pass LR filter (who makes just a stereo high-pass board?).

I could buy a board set, components for one - and just populate the high-pass section. Sheesh...now the system's a "cluster ....".
 
There is an expansion board for the minidsp 2x4, gives simple digital input and output over spdif and toslink, I recall. It's out of production, but still sold here and there. You could use that to make it a full duplex dsp device, and send the output to DACs you like.

But before you do that, are you finished with measurements, filter design and EQ design? Simply getting that right will change your perception of the minidsp greatly. I spent the day before yesterday building up serious EQ for my Fane 15-300TC fullrange units in backloaded horns. It was like a roller coaster, going from very promising to dull to better to almost broken to surprising to something I can live with. I was using a raspberry pi, software DSP (pulseaudio crossover rack, see the PC-based section of this site for details) and the Hifiberry Miniamp. I would have believed "OK" or bad, simply based on the price of the gear. But it can excell.
 
I've attached the MJ King spreadsheet of a design that I've built. It used minidsp, and I was happy with it. You can see that the LF cut off is 110Hz, and the HF cut off is 285Hz. Note also that the order of the crossovers is different. As you can see, the predicted frequency response of the system (on the penultimate page) is reasonably flat. I confirmed this with REW measurements.

Excellent documentation of your project!
 
Hi Ivo,

Thanks for your kind words. I suppose it is MJ King that deserves all/most of the credit. He makes open baffle (or U-frame/H-frame) quite easy.

Sadly, I no longer follow his latest work, he moved to facebook, which I'm not comfortable giving my personal info. If anyone's interested in his latest work, then the facebook links are here: Quarter Wavelength Loudspeaker Design

(Sorry if this is off-topic),
regards,
Ian
 
@Kaffiman - The Jolida amp is basically an LM1875 chip amp with a tube stage in front of it. There's a high voltage coupling cap between the tube stage and the LM1875. I imagine the two triodes per channel are arranged in a totem-pole topology, so I imagine the input impedance is whatever the volume control is. I've never drew out the schematic for it.

!
I had some issues using a tube pre amp, can you try to use it just from a computer or smartphone?
I understand that you are not very happy with your minidsp, and it's not exactly the best thing ever. Especially quirky when it comes to matching on inputs and outputs IMO.
But if you have a "good enough" pre-stage, even a headphone connector on a smartphone can work well, possibly better than a mid/high-ish output impedance tube amp.

There's a lot of reasons I've gone to use other DSP products than the basic minidsp, but it's not because of sound quality. I am happily using it in some systems.
 
@KaffiMann - so I measured the stereo output of the Aune and it looks like there's at least an 8K resistor in series with its internal output amplifier. That's driving a reputedly 6K input impedance of the miniDSP.

I would think these impedances would be just fine, but the Aune does sport a headphone output with a much stronger internal drive circuit. I could try it to see if the signal loading presented by the mini is the root cause issue.

Thanks!
 
how do the T-S parameters look for this driver? how consistent? evsentry3's pair had qts ~2 can't make sense of moving mass

ZRPCMGD.png