Hi there!
I am very very eager to make an EU round blind comparison of small fullranges.
The reason is that the work of xrk971 impressed me a lot, but... most of the drivers mentioned are not available in EU or are very pricey. Even the evergreen TC9 is ~21$ here 😡, but we have Faitals, Lavoces, Sicas and Visatons here on the cheap end and some scandinavian manufacturers on expensive.
What do I have:
-A room. It will be my living room, ~25sq. m, no problems with that
-Laptop with Windows. I would like to stay with free software
-Simple DAC (Apple usb->3.5 is a proper DAC)
-Aliexpress TDA3116 amp
-A little collection of ~3" fullrangers (photo attached)
-A will to make contribution to community
What I need to:
-Teach me, how its done... I have only vague idea...
-How to build or buy decent microphone
-To choose enclosure (-s). Will build it from MDF or foam
-Some aditional drivers. If someone from Europe will lend me their precious speakers - we all will be very gratefull. I think everyone is especially interrested in those which are less than 30$/unit here.
I have some musical education, the ears of mine can easilly distinguish apparent flaws, but only to a degree.
So, what the community think about this idea?
p.s. Sorry for my english
I am very very eager to make an EU round blind comparison of small fullranges.
The reason is that the work of xrk971 impressed me a lot, but... most of the drivers mentioned are not available in EU or are very pricey. Even the evergreen TC9 is ~21$ here 😡, but we have Faitals, Lavoces, Sicas and Visatons here on the cheap end and some scandinavian manufacturers on expensive.
What do I have:
-A room. It will be my living room, ~25sq. m, no problems with that
-Laptop with Windows. I would like to stay with free software
-Simple DAC (Apple usb->3.5 is a proper DAC)
-Aliexpress TDA3116 amp
-A little collection of ~3" fullrangers (photo attached)
-A will to make contribution to community
What I need to:
-Teach me, how its done... I have only vague idea...
-How to build or buy decent microphone
-To choose enclosure (-s). Will build it from MDF or foam
-Some aditional drivers. If someone from Europe will lend me their precious speakers - we all will be very gratefull. I think everyone is especially interrested in those which are less than 30$/unit here.
I have some musical education, the ears of mine can easilly distinguish apparent flaws, but only to a degree.
So, what the community think about this idea?
p.s. Sorry for my english
Attachments
Great idea! The details of the setup are explained in my threads. One thing you may want to add is a woofer to augment the bass and a crossover to use it as a FAST. In that way, the full range is only asked to work from 300Hz and above. MiniDSP is good to have but an analog line level opamp based XO could work well.
You also need a reference baffle that all the drivers can be fitted to. The baffle dimensions need to be the same.
I would recommend REW software and UMIK-1 mic for recording and measuring.
TPA3116 amp is fine.
Important thing is to keep everything identical except the driver under test. Then level match recording mic level to be identical. Louder recordings tend to be rated “better sounding”.
You also need a reference baffle that all the drivers can be fitted to. The baffle dimensions need to be the same.
I would recommend REW software and UMIK-1 mic for recording and measuring.
TPA3116 amp is fine.
Important thing is to keep everything identical except the driver under test. Then level match recording mic level to be identical. Louder recordings tend to be rated “better sounding”.
Last edited:
...Thanx for inspiration! Not everything I have understood, but it seems I need aditional:
-UMIK-1
-miniDSP or alternative
-some subwoofer? Which (cheap) subwoofer would you recomend? Please remember, that I am in EU, most of the cool drivers here are 2x the US price 🙁
My amp has USB input, so I maybe if I make some "dsp" in computer software, like just cut everything below 200-300Hz and play without subwoofer - will it be ok, or I still need a sub for better results?
You can use software like Audacity to play mono but split below 300Hz into left and above 300Hz to right channel into full range DUT. Basically a mono digital crossover and then use left and right channels of amp for woofer and fullrange. You need woofer for music to sound natural and good or else people can’t judge. Although it’s the full range they are judging. The reason you can’t ask the fullrange to make bass is each needs a specific cabinet to be optimal. Then it’s not a good comparison of the mid and top end.
There’s a thread where I tried this out - can’t remember where. But you get the idea.
Find any suitable 8in or bigger woofer with fairly clean flat response tonise for your bass unit. Design a basic vented box tuned to about 55Hz.
There’s a thread where I tried this out - can’t remember where. But you get the idea.
Find any suitable 8in or bigger woofer with fairly clean flat response tonise for your bass unit. Design a basic vented box tuned to about 55Hz.
Last edited:
Ok, understood. It will take some time, but the project is definitelly interresting.
Any recomendation for woofer? From my shallow experience FaitalPRO drivers are pretty good bang for the buck, maybe some woofer from them?
Any recomendation for woofer? From my shallow experience FaitalPRO drivers are pretty good bang for the buck, maybe some woofer from them?
Faital woofers are for pro use. They are expensive.
Try Visaton W200-8. Sealed 2.3 cu ft. Nothing simpler and you get 60Hz. Sealed bass is very clean and no issues with group delay.
Visaton W200-8 8" Woofer 8 Ohm
Try Visaton W200-8. Sealed 2.3 cu ft. Nothing simpler and you get 60Hz. Sealed bass is very clean and no issues with group delay.
Visaton W200-8 8" Woofer 8 Ohm
Yes, the price is nice and 2.3cu ft sealed is very doable and even WAF friendly after project
SO... YOU WANT TO DO IT RIGHT???
Be Careful my friend. This kind of driver comparison can be the worst Horse SH*t in the DIY speaker building realm.
Here are my tips on how to truly do a side-by-side comparison of drivers:
- completely transform/flatten (DSP is the only practical way) the frequency response of each driver, on axis, so that they are identical.
- Trim the SPL output of each transformed driver to within 0.3dB or better using gain adjustment
- Either use a high pass filter to eliminate the near-resonance output (by for example high passing all drivers at 300Hz) or use a Linkwitz Transform to make the behavior at resonance identical for all drivers.
- To conduct a proper blind test, no one participating in the test should know which driver is which, so you need to either blindfold the participants or put all the driver behind an acoustically transparent but visually opaque screen.
- You will need someone who is not participating in the test to, for example, randomly present one of the drivers to the testers, because no one can know which is which.
If you do all of that... Awesome. Your results will be VERY "believe-able". Otherwise... Meh. There is just too much influencing by the brain by cues of loudness, the appearance of the driver, the width of the passband, the driver's frequency response, etc. All of these will have an effect of the outcome unless they are eliminated as inputs for the test subjects. But all of these can and often are "corrected" using the crossover, so why not "correct" each driver to be identical and then see which has the better sound. At that point, most listeners (in my opinion) will have a very difficult time telling the difference between one driver and the next. What you are left with is the distortion signature, energy storage, resonances inherent to the driver, and the ability of the cone to stop transmission of the backwave.
It's a lot of work to do it "RIGHT"!!!
Last edited:
Thanx for advice! Yes, I know that this is a lot of work, but in couple of months I will be ready. A bit of clarification:
- By the definition it will be subjective testing - similar like X did in several rounds. We will try to select the best/nicest/most sweet sounding driver recorded in my own place with my own equipment, room, skills and etc. If someone would like to join - they are welcome, I am located in Vilnius, LT. Currently have 9 driver pairs, at least 7 of them are labeled as full-range units, and all of them are in 3 inch class. If someone would like to send me their own drivers - you are VERY welcome!
- There is very very very small chance that this will be a proper blind or double/tripple blind test as I will do it myself, probably alone, so I will know everything 😎 But it will be totally blind for everyone except me.
- At this moment have no idea, what is Likwitz Transform is, but if we are speaking about filtering <300Hz, it is a subwoofer job, right? Any recomendation for not expensive woofer, available in EU?
- I am very fresh in diy audio, but definitelly a quick learner and by education and work experience a bit an engineer. So, to do similar subjective testing like xrk971 did is pretty much possible. The only 2 obstacles are it will take some time and there is some financial cost - I will need to buy some equipment and to buy it wiselly.
- It is not hard for me to distinguish really good driver from properly bad one.
- This is subjective comarisson but with extra drivers, available in EU. We all will decide, which driver we like. It is a bit game, so lets have fun.
There is also not a science at all to subjectivelly have an opinion about any driver sound based on very personal hearing experience.
Last edited:
Charlie Laub,
Why would you EQ flatten the drivers response to be the same? That’s their intrinsic “voice” and what makes them sound different - the whole basis of the comparison. I agree that they should be level matched though.
You can do a blind test like I did here. It is not easy to tell which is which - people could only pick their favorite. And the favorite usually was the one with the flat response.
Subjective Comparison Threads
A Subjective Blind Comparison of 3in to 5in Full Range Drivers
A Subjective Blind Comparison of 3in to 5in drivers - Round 2
A Subjective Blind Comparison of 2in to 4in drivers - Round 3
A Subjective Blind Comparison of 2in to 4in drivers - Round 4
A Subjective Blind Comparison of 2in to 3.5in drivers - Round 5
Why would you EQ flatten the drivers response to be the same? That’s their intrinsic “voice” and what makes them sound different - the whole basis of the comparison. I agree that they should be level matched though.
You can do a blind test like I did here. It is not easy to tell which is which - people could only pick their favorite. And the favorite usually was the one with the flat response.
Subjective Comparison Threads
A Subjective Blind Comparison of 3in to 5in Full Range Drivers
A Subjective Blind Comparison of 3in to 5in drivers - Round 2
A Subjective Blind Comparison of 2in to 4in drivers - Round 3
A Subjective Blind Comparison of 2in to 4in drivers - Round 4
A Subjective Blind Comparison of 2in to 3.5in drivers - Round 5
... and then there's the problem of defining the target response: If you EQ to flat on axis, still the power response will be somewhat different, especially for widebanders which operate in breakup mode in the upper spectrum.
An introduction to reliable measurements from Floyd Toole: Objective Loudspeaker Measurements to Predict Subjective Preferences? | Audioholics
An introduction to reliable measurements from Floyd Toole: Objective Loudspeaker Measurements to Predict Subjective Preferences? | Audioholics
I would not EQ them. Curious to hear if these newer drivers sound more similar to each other compared to the ones tested a few years ago.
...There are some unbelievably bad sounding ones, which are not cheap either. You definitelly will hear the difference.
As I told project is alive, but it will take some time, as I am busy with other stuff. I also hope in month or two will get some Daytons also extra Visatons in the mix, no less than 11 pairs of drivers.
I am more interested in the distortion characteristics at various levels.
Which would definitely be different between the drivers, both with/without EQ.
I think that this is a great project and hope people are going to be bit more open to it. It’s all about subjective impressions and all we can do is our best to keep everything the same except the DUT (device under test).
Perhaps choose music that is mostly vocals and highs without much bass below 200hz and play these in a standard IEC open baffle to keep the naysayers at bay. It’s nice to have the bass unit for FAST but then our ears are convolving the bass and fullrange DUT.
Most importantly is to use identical everything except DUT. So the SPL in the recording must be matched via measurement.
Perhaps choose music that is mostly vocals and highs without much bass below 200hz and play these in a standard IEC open baffle to keep the naysayers at bay. It’s nice to have the bass unit for FAST but then our ears are convolving the bass and fullrange DUT.
Most importantly is to use identical everything except DUT. So the SPL in the recording must be matched via measurement.
I think that's a good argument for EQing them flat, any FR distortions would exacerbate the others?
Yup!
That's the idea.
And I'd like double music samples on everything.
One set of more relaxed music, and one set with a lot of stuff happ'nin.
For instance a sample using smooth jazz with vocal, and then another sample using some intense drum'n'bass take off kinda thing.
Sometimes easier to catch the differences when the speakers are "Stressed" a bit.
That's the idea.
And I'd like double music samples on everything.
One set of more relaxed music, and one set with a lot of stuff happ'nin.
For instance a sample using smooth jazz with vocal, and then another sample using some intense drum'n'bass take off kinda thing.
Sometimes easier to catch the differences when the speakers are "Stressed" a bit.
There must be some HiFi communities in Vilnius - with meetings and everything. Im sure such a group would be super-exited to participate in a blindtest. Such events are something that earns a lot of respect in HiFi circles.
Cheers!
Use any amplifier for driving the woofers. -Basicly I mean there is no need for an actual "subwoofer", but using a subwoofer would also be possible. I just think that making a single subwoofer sound right compared to stereo woofercabinets is not for beginners
Last edited:
- Status
- Not open for further replies.