Inspired by Matthew Morgan J and his 6th and 8th order superplanar designs and my own experiments with the TPQWR (ROAR) I got this idea:
It is a tapped pipe/compound horn hybrid where the short front-section greatly extends the usable passband compared to a normal tapped horn or tapped pipe.
Here is a simple comparison between a normal tapped pipe and a TPCH, both with the same Alpine SWR8D2 driver.
The TPCH8 has a usable passband of 27 - 160 Hz. It sounds great and has a very powerful low distortion clean output all the way up to we start hearing the Tamp E800 clip at 450 watts.
An externally hosted image should be here but it was not working when we last tested it.
It is a tapped pipe/compound horn hybrid where the short front-section greatly extends the usable passband compared to a normal tapped horn or tapped pipe.
An externally hosted image should be here but it was not working when we last tested it.
Here is a simple comparison between a normal tapped pipe and a TPCH, both with the same Alpine SWR8D2 driver.
The TPCH8 has a usable passband of 27 - 160 Hz. It sounds great and has a very powerful low distortion clean output all the way up to we start hearing the Tamp E800 clip at 450 watts.
An externally hosted image should be here but it was not working when we last tested it.
Here is the general layout.
An externally hosted image should be here but it was not working when we last tested it.
Two layers of 18 mm plywood makes a strong and stiff platform for the driver.
An externally hosted image should be here but it was not working when we last tested it.
It is standing beside a simple OD-QW with the same type of driver.
An externally hosted image should be here but it was not working when we last tested it.
Here is an actual measurement with my Umik 30 cm from the mouth - compared with the Hornresp simulation we used to design the TPCH8.
An externally hosted image should be here but it was not working when we last tested it.
The GD as measured and compared to the Hornresp simulation.
An externally hosted image should be here but it was not working when we last tested it.
THD measured outdoors at 3 meters and approximately 100 watts.
Quite impressive for a single 8 inch driver.
YouTube
My 15 year old son built this sub for our car. We are both very impressed by the sound quality and the quantity. The sound pressure is intense inside our car. I could never imagine that a single 8 inch driver could play this loud and clean.
What's the dimensions of the box?
How do you access the driver after the box is sealed up?
The vent looks partly blocked in that video of the box in the car. Did that effect its response?
A 30 cm distance from the vent seems a bit too far for an in-house measurement. You've probably got a bit of the room's effect in that response curve.
How do you access the driver after the box is sealed up?
The vent looks partly blocked in that video of the box in the car. Did that effect its response?
A 30 cm distance from the vent seems a bit too far for an in-house measurement. You've probably got a bit of the room's effect in that response curve.
92 x 47,2 x 29,6 cm if I remember correctly.
My son wanted to install the driver with the magnet protruding into the short midbass pipe, which means that we have to cut out the driver with a tiger saw if we need to replace it. I protested against this, but it was his project, so he got the final word.
There was a lot of open air around the side of the mouth that is not easily seen in the video. We did not hear any signs of port-turbulence or port compression, so I don´t think it was a problem.
An externally hosted image should be here but it was not working when we last tested it.
Here is an outdoor measurement with the microphone 20 cm from the mouth (purple line) compared to a Hornresp-sim.
No smoothing.
Cheers,
Johannes
Hmm... ain't no way the Hornresp sim, if it's an accurate one is going to be LESS sensitive than the actual measurement at low frequencies below the lowest resonance frequency, particularly at what seems to be almost 5dB less sensitive. Looks like the purple line needs to be adjusted about 5dB downwards.
Try engaging the "Lossy Le" function to see if it brings the sim closer to the measured response.
I did not try to measure at the same power level as the Hornresp simulation. I just tried to make them reasonably similar so I could compare the shape of the passband.
It is cold outside today, so I did not want to waste any time adjusting stuff.
A picture from when I was measuring the response at 1 meter.
Cheers,
Johannes
It is cold outside today, so I did not want to waste any time adjusting stuff.
An externally hosted image should be here but it was not working when we last tested it.
A picture from when I was measuring the response at 1 meter.
Cheers,
Johannes
PARAFLEX
Johannes,
Good to see you experimenting with this type of cabinet!
This is what we would call a PARAFLEX (but we usually implement a little bit of expansion in the main path, primarily the S45 segment , it can still be "stepped" expansion as the ROAR series has) .....
We consider the Paraflex a variant of the Super Planar series ....
Using some expansion in the last segment of the main path can improve output & efficiency and lower Group Delay .. . This is the case with many drivers , especially car audio style drivers ..
Also, in order to have your Hornresp model match your measured results more accurately you would need to add some S34 length to your model , approximately half the width of your high-tuned resonator (from the side view) ... . You will notice that doing this increases the Group Delay in the model ...
In some cases you can flatten the response slightly by adding just a little bit of driver offset in the main path, just a small amount like 20cm or 30cm, this will help to reduce the "saddle" shape in the response in some cases but not others .... May help reduce the prominence of the peak near the fundamental .... Not a huge improvement but an improvement nonetheless .. .
When you are dealing with drivers that don't have remarkably strong motors in relation to the driver's cone area (also taking lossy inductance into account)
we have found that it is often best to go with a high-tuned resonator that is 10cm to 20cm shorter and giving it considerably more cross sectional area (CSA) making our high-tuned resonator more like a cavity resonator that shares volume with the rest of the cabinet, this helps to concentrate more of our energy below 100hz thereby improving average output in the range where most people are actually using their subwoofers and this method also often lowers the group delay near the cabinet's tuning ... .
The aggressive rising response can be very useful in a few different practical situations
Both in PA and Car Audio ...... For example: In car audio where you need to compensate for the vehicle's cabin gain this works out very well! In PA applications the aggressive rising response is very helpful when coupling many of these subs together in an array, and response flattens, much like what is observed when coupling many FLH cabinets together .... Verify by using the "Multiple Speakers" feature in Hornresp ....
I know this is a lot of information ... . Just trying to help .... We have a lot to offer on this arrangement because we have been highly focused on the Paraflex for a while now... Most of the discussion takes place in the facebook group but i update the DIYaudio discussion when i can ... Compound loading 6th order quarterwave "Super Planar" horns and pipes concepts/builds ....
Many Paraflex cabinets have been built for Car Audio so far here in the states , and we have some friends working on some PA builds ......... Johannes, You are the first to build a Paraflex in all of Europe as far as i am aware! But there are a few people in Europe with projects in the works for the near future so your timing was good ... GKH in Austria has had the plans to build a Paraflex for his 15LB100 drivers for several months or more now but i don't know if he ever got around to it , I haven't heard back from him on that yet ..
Here is a list of various Paraflex Layouts that we have come up with ..... Only 3 or 4 of these layouts have been employed so far , because they were most appropriate, but it is good to have options ... Notice how most of these have some expansion in the latter part of the main (LF) path ..... These layouts are sketched with a 15" or 18" loaded in them, so the proportions are much different when loading with an 8" or a 10" or multiple such small drivers .. .
NOTE: In a scenario where you have a driver with high free-air resonance (Fs) value and low Vas figure you can run into problems with unacceptably high Group Delay when trying to tune the cabinet very far below the driver's Fs ..... ..... This isn't necessarily what is happening with the Alpine 8 , but it has been something that we have run into in the past , and can be an issue with resonant cabinets in general .. .
You may find that it is easier to get the Paraflex to behave well with multiple Alpine 8s in a single Paraflex cabinet .. Or multiple 10s .. . Or just a single larger driver ....Meaning more cone area...... Generally drivers with lower Qes/Qts values & plenty of effective motor force are preferable along with an Fs value which is neither too low or too high in relation to your desired cabinet tuning ...
Johannes, please feel free to ask if you ever want any help optimizing the Paraflex type layouts ... I will gladly assist .. As long as suitable drivers are available we can make great things happen 
Johannes,
Good to see you experimenting with this type of cabinet!
This is what we would call a PARAFLEX (but we usually implement a little bit of expansion in the main path, primarily the S45 segment , it can still be "stepped" expansion as the ROAR series has) .....
We consider the Paraflex a variant of the Super Planar series ....
Using some expansion in the last segment of the main path can improve output & efficiency and lower Group Delay .. . This is the case with many drivers , especially car audio style drivers ..
Also, in order to have your Hornresp model match your measured results more accurately you would need to add some S34 length to your model , approximately half the width of your high-tuned resonator (from the side view) ... . You will notice that doing this increases the Group Delay in the model ...
In some cases you can flatten the response slightly by adding just a little bit of driver offset in the main path, just a small amount like 20cm or 30cm, this will help to reduce the "saddle" shape in the response in some cases but not others .... May help reduce the prominence of the peak near the fundamental .... Not a huge improvement but an improvement nonetheless .. .
When you are dealing with drivers that don't have remarkably strong motors in relation to the driver's cone area (also taking lossy inductance into account)
we have found that it is often best to go with a high-tuned resonator that is 10cm to 20cm shorter and giving it considerably more cross sectional area (CSA) making our high-tuned resonator more like a cavity resonator that shares volume with the rest of the cabinet, this helps to concentrate more of our energy below 100hz thereby improving average output in the range where most people are actually using their subwoofers and this method also often lowers the group delay near the cabinet's tuning ... .
The aggressive rising response can be very useful in a few different practical situations
Both in PA and Car Audio ...... For example: In car audio where you need to compensate for the vehicle's cabin gain this works out very well! In PA applications the aggressive rising response is very helpful when coupling many of these subs together in an array, and response flattens, much like what is observed when coupling many FLH cabinets together .... Verify by using the "Multiple Speakers" feature in Hornresp .... I know this is a lot of information ... . Just trying to help .... We have a lot to offer on this arrangement because we have been highly focused on the Paraflex for a while now... Most of the discussion takes place in the facebook group but i update the DIYaudio discussion when i can ... Compound loading 6th order quarterwave "Super Planar" horns and pipes concepts/builds ....
Many Paraflex cabinets have been built for Car Audio so far here in the states , and we have some friends working on some PA builds ......... Johannes, You are the first to build a Paraflex in all of Europe as far as i am aware!
But there are a few people in Europe with projects in the works for the near future so your timing was good ... GKH in Austria has had the plans to build a Paraflex for his 15LB100 drivers for several months or more now but i don't know if he ever got around to it , I haven't heard back from him on that yet .. Here is a list of various Paraflex Layouts that we have come up with ..... Only 3 or 4 of these layouts have been employed so far , because they were most appropriate, but it is good to have options
... Notice how most of these have some expansion in the latter part of the main (LF) path ..... These layouts are sketched with a 15" or 18" loaded in them, so the proportions are much different when loading with an 8" or a 10" or multiple such small drivers .. .An externally hosted image should be here but it was not working when we last tested it.
NOTE: In a scenario where you have a driver with high free-air resonance (Fs) value and low Vas figure you can run into problems with unacceptably high Group Delay when trying to tune the cabinet very far below the driver's Fs ..... ..... This isn't necessarily what is happening with the Alpine 8 , but it has been something that we have run into in the past , and can be an issue with resonant cabinets in general .. .
You may find that it is easier to get the Paraflex to behave well with multiple Alpine 8s in a single Paraflex cabinet .. Or multiple 10s .. . Or just a single larger driver ....Meaning more cone area...... Generally drivers with lower Qes/Qts values & plenty of effective motor force are preferable along with an Fs value which is neither too low or too high in relation to your desired cabinet tuning ...
Johannes, please feel free to ask if you ever want any help optimizing the Paraflex type layouts
... I will gladly assist .. As long as suitable drivers are available we can make great things happen
Last edited:
Gerald ,
I hope you are feeling better! 5 weeks is a long time to be down
..... Many of my friends were sick...I am glad that everyone i know recovered ... The 2017-2018 flu season is said to have been especially nasty because of the strain that has been circulating around, but thankfully the rate is slowing down now as of last month (according to official reports) ...
I forgot to ask, how does its output compare to the OD-QW build?
Do you have comparable FR measurements for both builds?
Thanks for all the information! I am sorry for my very delayed answer. I have been down in a hellhole of extreme fatigue due to my ME/CFS, but I am getting a little better again.
I saw my "TPCH" as a variation of your Paraflex due to the constant cross section pipes and a shared mouth. I have missed that you use a shared mouth in several of your new designs. I remembered the dual mouth compound horn designs, so I thought this was a new variation to your Paraflex.
I did not build this thing. It was my 15 year old son who wanted some quality bass in our car.
We tested this design at a local car audio meet, and my son is very happy by his first ever car audio bass spl measurment.
An externally hosted image should be here but it was not working when we last tested it.
With 300 - 400 watts and a single Alpine type R 8 inch driver.
We guesstimate that there is at least 5 dB more lurking deep down in the pipe-sections. It sounds very clean and powerful at 134,8 dB. No hint of compression of turbulence despite the cramped mouth exit. We had installed the bass the night before the car audio meet, so we had not had the time to adjust and optimize anything. It was just the raw "set everything half way just to get a first measurement" -test.
We have a 2000 watt amp, but we did not want to fry the voice-coil as the very first thing we did.
Everyone who hears it is very impressed. It is powerful and clean, with both a very deep response and a great midbass punch.
Cheers,
Johannes
Johannes, Glad to provide ..... If you would ever like any help optimizing a Paraflex model in Hornresp, just let me know
We have found that some drivers really excel in the design (driver choice is the critical first step) ...... . . We have been calling the compound horn design with dual mouths The Super Planar or Super Planar 8th (non shared mouths , no path merge) ............
Then the compound horn versions with the merged path and shared mouth are being called Paraflex.
The original code name for the Paraflex was "Badass Bandpass" ..
. Hehehe.. That was from a few years ago and was just theory back then..... Finally we saw the first of them being built around the end of last year with Dustin Morgan's 1x12 Paraflex cabinet and Matt Kimble's 1x15 Paraflex cabinet being the first two ....
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.