Nice work, Plott!
You could use a HyperSET board for the tube buffer of course. 😀
Then the safe turn on logic sequence would be all done for you by connecting the the 3 pin connector shown here to the 1200AS1 board. It’s the enable feature on the 1200AS that mutes the output until the buffer is warmed up.
You could use a HyperSET board for the tube buffer of course. 😀
Then the safe turn on logic sequence would be all done for you by connecting the the 3 pin connector shown here to the 1200AS1 board. It’s the enable feature on the 1200AS that mutes the output until the buffer is warmed up.
On the bench tonight is a passive filter for a friend’s single driver fullrange speaker. He built it with a repurposed bamboo curio box and a Dayton PS95-8 driver and some Aliexpress passive radiators in the back.
I measured it and simulated a filter in Xsim and arrived at a BSC using 1mH and 7ohm resistor followed by a 0.33mH and 2.2uF 2nd order low pass to tame the rising response. It measures really well - almost monitor like now and with a slight down tilt. If anyone wants to use this filter - make the speaker cabinet about 6in wide to match the baffle step correction. Otherwise the speaker dimensions are up to you. I think it’s a very worthwhile filter as parts are about $35 for the pair of speakers. Great sounding and very neutral with maybe a bit more presence in vocals with the small 1.5kHz bump.
Here is the filter schematic in Xsim:
Predicted Freq response vs the measured raw response (yellow):
Predicted Impedance:
Measured response (blue) vs raw (red):
Measured harmonic distortion (which improved with the filter):
Here is the temporary Wago connector P2P filter:
I assembled a second one and listened to it in stereo. It sounds very nice. Great midrange and very good sparkle without being fatiguing. Superb imaging as expected from a single driver fullrange setup. Here’s a video sample of the speaker - played at 2.0Vrms and about 7 ft away.
I measured it and simulated a filter in Xsim and arrived at a BSC using 1mH and 7ohm resistor followed by a 0.33mH and 2.2uF 2nd order low pass to tame the rising response. It measures really well - almost monitor like now and with a slight down tilt. If anyone wants to use this filter - make the speaker cabinet about 6in wide to match the baffle step correction. Otherwise the speaker dimensions are up to you. I think it’s a very worthwhile filter as parts are about $35 for the pair of speakers. Great sounding and very neutral with maybe a bit more presence in vocals with the small 1.5kHz bump.
Here is the filter schematic in Xsim:
Predicted Freq response vs the measured raw response (yellow):
Predicted Impedance:
Measured response (blue) vs raw (red):
Measured harmonic distortion (which improved with the filter):
Here is the temporary Wago connector P2P filter:
I assembled a second one and listened to it in stereo. It sounds very nice. Great midrange and very good sparkle without being fatiguing. Superb imaging as expected from a single driver fullrange setup. Here’s a video sample of the speaker - played at 2.0Vrms and about 7 ft away.
Last edited:
Thanks X!
Yes, I know - but I'm building two monoblocks and I want to try to design a simple input stage
I hope I don't fail.
I haven’t looked at that aspect yet. But, Qts and Fs are almost an exact match. The Vas in SB23 is 91 liters vs 57 liters on the RS225. Based on this, I think it should work fine since the volume of the TL is large and will not be an issue with the suspension compliance. The main difference will be the SB23’s less than flat response at the higher frequencies. So I think the crossover will have to be a higher order low pass.
Super deal on 0.53x Karlsonator cabinets by Jhofland available here.
https://www.diyaudio.com/community/...ator-cabinets-and-drivers.410584/post-7633792
https://www.diyaudio.com/community/...ator-cabinets-and-drivers.410584/post-7633792
Finished assembling the two cabinets with SB23’s and PS95’s. Currently bi-wired with two sets of binding posts for purposes of crossover development.
Inside of cabinet is stuffed with pink fiberglass. Inside of Dagger is 2/3 stuffed with fiberglass and last 1/3 with polyfill fluffed up.
Inside of cabinet is stuffed with pink fiberglass. Inside of Dagger is 2/3 stuffed with fiberglass and last 1/3 with polyfill fluffed up.
Something just did not sound right and I found out that I totally had the acoustic offset completely wrong. Once that was fixed, I went to make the XO point near 1kHz so that it can be a FAST (low frequency crossover). This ended up taking another inductor to make the woofer filter a higher order. The capacitor sizes also went up. This is not a simple or inexpensive crossover because the PS95-8 requires so many parts to make it behave well. But the payoff is a sizzle and sparkle at the top end like a dome tweeter. The SB23 gives a very high sensitivity and overall system is about 86dB (after baffle step) and -3dB is about 51Hz - which is great for a sealed box.
Best part is that it sounds GREAT! You can tell in 45 seconds that it is a great sounding speaker. Very musical and engaging. I’m happy to say that the XO complexity is worth it.
Measurement of actual XO behavior:
There is even a slight Harman House Curve (slight down tilt to the right) to prevent fatigue.
Best part is that it sounds GREAT! You can tell in 45 seconds that it is a great sounding speaker. Very musical and engaging. I’m happy to say that the XO complexity is worth it.
Measurement of actual XO behavior:
There is even a slight Harman House Curve (slight down tilt to the right) to prevent fatigue.
Second channel XO now assembled.
Closeup of PS95-8 on the baffle:
Closeup of SB23NRXS40-8:
Now playing in stereo:
In the end, the final voicing test led me to pad the tweeter by an extra 1ohm to 3R at R1. This gave a very well balanced and less forward presence in the 1kHz to 3kHz range while keeping the highs above 10kHz the same.
Here is a video of various tracks with the above XO. It’s a long video but there are a lot of songs of all different genres. I found that the bass was surprisingly deep and plentiful for a sealed cabinet. Enjoy!
Closeup of PS95-8 on the baffle:
Closeup of SB23NRXS40-8:
Now playing in stereo:
In the end, the final voicing test led me to pad the tweeter by an extra 1ohm to 3R at R1. This gave a very well balanced and less forward presence in the 1kHz to 3kHz range while keeping the highs above 10kHz the same.
Here is a video of various tracks with the above XO. It’s a long video but there are a lot of songs of all different genres. I found that the bass was surprisingly deep and plentiful for a sealed cabinet. Enjoy!
On the bench tonight is using a hole saw to add a 60mm dia reflex vent to see if the bass response has better response.
The sealed box actually seems to have deeper bass extension. This cabinet is 24L and is too small for the 94L Vas of the SB23 for an ideal bass reflex cabinet. This was the prediction of 60mm x 216mm long round vent:
The sealed box actually seems to have deeper bass extension. This cabinet is 24L and is too small for the 94L Vas of the SB23 for an ideal bass reflex cabinet. This was the prediction of 60mm x 216mm long round vent:
Best to see the effect of port is to do impedance sweep by dats. Before and after. Or open vs closed.
In room low fr response is not accurate.
Here is DATS sweep showing box bass reflex vent tuning is at 30Hz - close to the driver's free-air Fs. Sealed box tuning is 52.4Hz and matches the measurements. The DATS revealed that I have too low of an impedance dip near 2kHz so the tweeter filter 0.2mH shunt to ground is shorting out essentially. I will need to work on that.
I will have to do a closeup mic measurement of woofer and vent output and estimate the combined output. The room effect indeed will make this hard to see as room modes dominate and it depends where speaker is placed.
I will have to do a closeup mic measurement of woofer and vent output and estimate the combined output. The room effect indeed will make this hard to see as room modes dominate and it depends where speaker is placed.
Last edited: