Hand crafted SP with ONKYO "OM-OF101" vol.6 - finale

I put into action the addition of a tweeter to increase the high frequency energy of this loudspeaker.



First, I checked the 30-degree characteristic of the OM-OF101 unit alone again, and found that it gently dropped from above 4KHz.

- Frequency response of OM-OF101 full range (30 degrees)


If the OM-OF101 unit was to be connected to a tweeter, I felt that an 8k-12kHz crossover was appropriate.

The additional tweeter was an Olasonic super tweeter included in DigiFi No.20 magazine.
I got it about a year ago at a bargain price, and it had been forgotten in a corner of my bookcase, so let's have it sing this time.

First, I tested the frequency response of the super tweeter only, using different capacitors that I have.

There were four types of capacitor capacitance: 0.22μF, 1.0μF, 1.5μF, and 3.3μF. The manufacturers varied, and included some that had been used quite some time ago. I didn't take a picture of them because they looked bad with butyl rubber stains and such.

To match the 30 degree response of the OM-OF101, I measured the frequency response at a 30 degree angle from the front axis of the Olasonic tweeter.

- Frequency response of Olasonic tweeter alone (30 degrees, 0.22μF)


- Frequency response of Olasonic tweeter alone (30 degrees, 1.0μF)


- Frequency response of Olasonic tweeter alone (30 degrees, 1.5μF)


- Frequency response of Olasonic tweeter alone (30 degrees, 3.3μF)


The 0.22μF capacitance had a low high frequency level. The 1.0μF capacitance extended to 16kHz. The 1.5μF capacitance had a lower level above 10kHz due to the phase. I thought that the 1.0μF capacitance was the safest choice.

Next, I measured the frequency response on the axis. I retreated from the baffle surface by 0 mm / 15 mm / 20 mm.

- Frequency response of OM-OF101+Olasonic tweeter (on axis, 00mm)


- Frequency response of OM-OF101+Olasonic tweeter (on axis, -15mm)


- Frequency response of OM-OF101+Olasonic tweeter (on axis, -20mm)


The offset of 20mm was quite low from 10kHz onwards. In the case of the 00mm offset, which is the same as the baffle surface, it seemed to go down gradually from 4kHz.

The offset of 15mm showed an increase in the high frequency range. This was because the voice coil position of the OM-OF101 unit was around 12-15mm, so I guessed it was standard practice to match the voice coil position. After that, I decided to fine-tune the position while listening.

I could also try changing the polarity, but I decided to play a few songs in this state (C=1.0μF, 15mm from the baffle surface, normal phase) to check the sound.

At first listen, the reproduced sound was clear and crisp in the high frequency range and the center position was clear. Because of the super tweeter, the connection between the mid and high frequencies was natural.

The female vocalist sang without any sense of clutter. The bass line in the low range also seemed to have changed to a tighter direction. This might have been an effect of aging the unit.

Finally, I compared the characteristics with those of the ScanSpeak 10F/8422-03 10cm full-range unit, which was included in the August 2012 issue of Stereo magazine.



I designed this box with a 3.7L capacity and bass reflex rear duct fd=108Hz. The box is designed to be installed horizontally, so I felt it would be a bit uncomfortable if it was installed vertically.

- Frequency response of ScanSpeak 10F/8422-03


- Frequency response of ONKYO OM-OF101


Their sound pressure levels seemed to be comparable, but this box had a significant drop below 100Hz. There was a 9 year difference between the two boxes, so it was hard to compare them under the same conditions.
.s

Apple "HomePod mini" - Review and frequency response measurement.

I just got the Apple "HomePod mini" as an AirPlay evaluation device. The box was smaller than I had imagined.



I immediately took it out of the box and tried to set it up, but it took quite a bit of time because I had deleted the Home app or turned off Bluetooth.
After I managed to finish the setup, I can play music from Apple Music app without any problem.

So I decided to measure the frequency response of this device by playing sound from iTunes.

It was measured on a table at a distance of 1 meter with "Audio Frequency Analyzer" (AFA) app.

I played a sine wave at 1kHz to roughly adjust the playback level.
Then I played back pink noise and measured the frequency response by AFA app, holding the peak for about 15 seconds and doing a 1/3 octave band analysis.

- FFT measurement (playback level adjustment: 1m, Sine wave 1kHz)


- Frequency response measurement (1m, Pink noise, 1/3 octave band analysis, HOLD on)


In this measurement, the levels at 160Hz and 200Hz were a little low, probably due to standing waves, as they also appeared in other speaker measurements.

For the size of the enclosure, the bass extended well down to 63Hz. The high frequency range above 2kHz was about 2 to 4dB lower than the mid range, and the characteristic sound point seemed to be around 12kHz.

The auditory response was close to this frequency response, and the bass feeling was more than I expected, so it was easy to listen to music even at low volume. I felt that the drivers and passive radiator were well designed for this small size.

The HomePod mini does a very good job of playing music for background music!
.s

Hand crafted SP with ONKYO "OM-OF101" vol.5 - Extensions

This gimmick is a duct plug!

Cut from a scrap piece of OSB plywood.



I had to cut it out six times in total to fit the diameter of the duct holes. The thing behind it is the auxiliary plate for the tweeter, which is in the process of being painte

Here is a comparative image of bass reflex mode and sealed mode.

- Bass reflex mode


- Sealed mode.


I immediately measured the frequency response of the bass reflex mode and the sealed mode.

They didn't seem to have any extreme differences, but the 63Hz level seemed to be a little lower in the sealed mode.

- Frequency response in Bass reflex mode


- Frequency response in Sealed mode


As a 10.5L sealed box, f0c (f0 of the unit when installed in the box) is calculated around 101Hz, which is a little higher frequency than bass reflex.

My impression is that the bass is a bit suppressed in sealed mode, and bass reflex mode is probably the better choice.

If sealed mode is used, it would be better to add a little more sound absorbing material.
.s

Hand crafted SP with ONKYO "OM-OF101" vol.4 - Assembly completion and evaluation

The summer heat has subsided and the mornings and evenings are becoming more pleasant.





Get the faston terminals, and the speaker box for the OM-OF101 is complete!



The size image of the box looks like this. The magazine is B5 size.



I installed the SP unit in the completed box and played music at a low volume to check the connections.



Polarity is OK with no unusual noises. First, I measured the frequency response of the left and right channels.



I used the "Audio Frequency Analyzer" to measure the frequency response of the pink noise playback, 1/3 octave band, after matching the level with a 1kHz sine wave.

- frequency response L-ch (Initial)


- frequency response R-ch (Initial)


It is wonderfully flat.

There is no extreme difference between the left and right channels in the low frequency range, so the SP box assembly seems to be fine. The low frequency levels at 63/80Hz are the same, so the duct tuning seems to be good.

Anyway, I listened to some songs in this state.
Instrumental music, female vocals, orchestra, quartet,... I could feel the rhythm of the bass well enough. And the separation of the instruments was sufficient, and the sense of localization was fine.

However, I felt that the sound was lacking something. The vocals did not come forward. I wonder if this is a matter of personal preference...

So I tried reducing the sound absorbing material a little.

- frequency response (reducing the sound absorbing material)


As a result, the 63Hz level dropped a little. The overall impression was the same, although I felt that the volume of the bass might have been reduced in the auditory sense as well.

At this time, the distance between the left and right speakers was about 130cm, which was a little too wide for a 10cm unit.

I narrowed the distance to 90cm and listened again. I felt that the vocals became a little clearer.

So, I measured again for confirmation.



- frequency response (On the axis 100cm, 0 degree)


I measured again at a different angle, 1m above the axis. I attached a 1m long string to the tripod leg and moved it to a 30 degree position.

By using "ClinoScope", I found the angle in one shot just by turning the iPhone.



The frequency response at 30 degrees showed a gentle drop in the frequency range higher than 5 kHz.

- frequency response (On the axis 100cm, 30 degrees)


The reason for the unsatisfying sound may be the high frequency characteristics, or may be a matter of the sound directivity when the space between the speakers is too wide.

I tried to measure the 60-degree response as well, but the sofa was in the way. I had no choice but to measure at 45 degrees for reference.

- frequency response (On the axis 100cm, 45 degrees)


As for this unit design, I felt that the emphasis would be on the low-mid range.

It reproduced enough bass feeling even with a simple bass reflex, and almost no unpleasant sound came out. I would have liked a little more sharpness, but that's a matter of personal preference.

As mentioned in the article on box construction in the magazine, this may be the reason why a box with moderate resonance may be more suitable.

The fact that it is not flavored with a specific frequency range means that it is a seriously designed unit.

It will be interesting to use it as a midrange speaker unit in a 3-way system.

As a future plan, I'm thinking of putting a tweeter on it.
.s

Hand crafted SP with ONKYO "OM-OF101" vol.3 - painting

The temperature has dropped in September, and it feels a bit like autumn.

While I was waiting for the purchase of parts, I varnished the speaker box. (Really be satisfied with the way it looks?)



I lightly painted the back plate with black spray. This was only one coat of paint, so it was done in an instant.
　　


The only remaining step is to install the terminals and the unit.
.s