Hi everyone,
I'm actually motivate to build a new pair of speakers. My recent reading from the old L'Audiophile magazine make me developed a special interest for the petite Onken as described by Jean Hiraga (Audiophile n. 25, september 1982):
http://www.asrr.org/biblioteca/Revue%20 ... ONKEN.html
Considering the Altec 414-8B associated with them, do you guys think it's better to find a used pair or buy a 414-8B reedition from Great Plain Audio?
http://www.greatplainsaudio.com/downloa ... 0Sheet.pdf
At this time, the main challenge for me seems to decide with which horn driver I will design it. A member on French forum Elektor use a TAD TD2001 in a two ways design, a design which I like. But the TAD is VERY expensive. So if some of you could suggest other options, I'm willing to read them. I imagine that the cross-over will held at around 650 Hz. So, I'm looking for a horn driver with a good linearity between 500 Hz to 20 000 Hz.
Thank you,
Sébastien
I'm actually motivate to build a new pair of speakers. My recent reading from the old L'Audiophile magazine make me developed a special interest for the petite Onken as described by Jean Hiraga (Audiophile n. 25, september 1982):
http://www.asrr.org/biblioteca/Revue%20 ... ONKEN.html
Considering the Altec 414-8B associated with them, do you guys think it's better to find a used pair or buy a 414-8B reedition from Great Plain Audio?
http://www.greatplainsaudio.com/downloa ... 0Sheet.pdf
At this time, the main challenge for me seems to decide with which horn driver I will design it. A member on French forum Elektor use a TAD TD2001 in a two ways design, a design which I like. But the TAD is VERY expensive. So if some of you could suggest other options, I'm willing to read them. I imagine that the cross-over will held at around 650 Hz. So, I'm looking for a horn driver with a good linearity between 500 Hz to 20 000 Hz.
Thank you,
Sébastien
Faital Pro LTH142 horns and one of their 1.4" drivers. Their compression drivers are supposed to be really really good. One such iteration at
Cornscala Style D | Critesspeakers.com
Crites refer to the TAD TD4002 in the text...
an other scaled down to 12" and 1"
Beyma 12″+Faital 1″+horn |
There are more places around the net with people that have good things to say about those horns and drivers.
With a sharp active crossover analog or digital the horns can be used quite low in a domestic setting. If you use a passive crossover you have to be more careful.
Cornscala Style D | Critesspeakers.com
Crites refer to the TAD TD4002 in the text...
an other scaled down to 12" and 1"
Beyma 12″+Faital 1″+horn |
There are more places around the net with people that have good things to say about those horns and drivers.
With a sharp active crossover analog or digital the horns can be used quite low in a domestic setting. If you use a passive crossover you have to be more careful.
Hi DrBoar, hi everyone,
I think that I found an interesting path with some Radian compression drivers. Some folk comment that the 475PB is pretty similar to a TAD TD2001, but for a far cheaper price.
Plus, some other Radian also present a good linear frequency response from 600 Hz to 20kHz.
That's the case with the Radian 745PB and the 850PB.
Sébastien
I think that I found an interesting path with some Radian compression drivers. Some folk comment that the 475PB is pretty similar to a TAD TD2001, but for a far cheaper price.
Plus, some other Radian also present a good linear frequency response from 600 Hz to 20kHz.
That's the case with the Radian 745PB and the 850PB.
Sébastien
Greets!
Unless you or a knowledgeable friend knows the history of a pair and they are original used in a HIFI app or at least recently been re-coned, remagnetized by GPA with documented T/S measurements, then buy new as there’s too many ignorant and/or disreputable folks trying to capitalize on the Altec name and even GPA has been known to do way out of spec re-cones.
AFAIK though, GPA’s new drivers have been very close to spec [or arguably better in some cases] and when adjusted for inflation, no more expensive than the originals, so all things considered, I consider them a relative bargain by today’s standards.
That said, plugging its published specs into Hiraga’s formula yields a 106.91L/40.14 Hz alignment, so ~1/3rd smaller and tuned a bit lower than his 155 L/45 Hz, so some cab fine tuning may be required.
WRT horn/driver choice, I don’t know what the original mini-Onken used, but the ‘sweet’ spot for a wide BW 12” in general for a 90 deg horn [-6 dB] is ~1356 Hz and ~1500 Hz for the 414 in particular based on their own polar mapping of various 12” drivers, so to my way of thinking, a lower XO point is kind of a waste of much of the 414’s ‘magic’.
At ~1 kHz, its response is still ~ ‘setting’ on the baffle [~160 deg/-6 dB], so any lower XO point is basically just a waste of a pair of excellent wide BW drivers, not to mention limits one’s HF driver choices.
I queried the late JMMLC some years ago WRT the Altec 802-8G [the best overall performing 1” driver I’ve used] Vs the TAD TD-2001 Beryllium after he had mentioned doing some comparisons on his own system and he considered the latter’s better HF performance differences were subtle at best, so I concluded that it’s probably a waste of $$$ unless ones hearing is still youthful/undamaged.
Note that this was before GPA released its Series II version though, which if its published response plot is accurate, maybe reduces the already subtle differences to the point where only a child’s or young adult female’s keener mids/HF hearing might make the extra cost worthwhile.
Assuming at least a 1 kHz XO, DIYing an original ~212 Hz Unity WG [less mids holes, mounting of course] would be my choice or at least a 500 Hz expo horn.
GM
Unless you or a knowledgeable friend knows the history of a pair and they are original used in a HIFI app or at least recently been re-coned, remagnetized by GPA with documented T/S measurements, then buy new as there’s too many ignorant and/or disreputable folks trying to capitalize on the Altec name and even GPA has been known to do way out of spec re-cones.
AFAIK though, GPA’s new drivers have been very close to spec [or arguably better in some cases] and when adjusted for inflation, no more expensive than the originals, so all things considered, I consider them a relative bargain by today’s standards.
That said, plugging its published specs into Hiraga’s formula yields a 106.91L/40.14 Hz alignment, so ~1/3rd smaller and tuned a bit lower than his 155 L/45 Hz, so some cab fine tuning may be required.
WRT horn/driver choice, I don’t know what the original mini-Onken used, but the ‘sweet’ spot for a wide BW 12” in general for a 90 deg horn [-6 dB] is ~1356 Hz and ~1500 Hz for the 414 in particular based on their own polar mapping of various 12” drivers, so to my way of thinking, a lower XO point is kind of a waste of much of the 414’s ‘magic’.
At ~1 kHz, its response is still ~ ‘setting’ on the baffle [~160 deg/-6 dB], so any lower XO point is basically just a waste of a pair of excellent wide BW drivers, not to mention limits one’s HF driver choices.
I queried the late JMMLC some years ago WRT the Altec 802-8G [the best overall performing 1” driver I’ve used] Vs the TAD TD-2001 Beryllium after he had mentioned doing some comparisons on his own system and he considered the latter’s better HF performance differences were subtle at best, so I concluded that it’s probably a waste of $$$ unless ones hearing is still youthful/undamaged.
Note that this was before GPA released its Series II version though, which if its published response plot is accurate, maybe reduces the already subtle differences to the point where only a child’s or young adult female’s keener mids/HF hearing might make the extra cost worthwhile.
Assuming at least a 1 kHz XO, DIYing an original ~212 Hz Unity WG [less mids holes, mounting of course] would be my choice or at least a 500 Hz expo horn.
GM
Hi GM, hi everyone,
GM, your post is of great interest and contains a lot of informations. Meanwhile, I had the chance to talk to a member on Elektor which build the petite Onken and he told me that in its design, the Altec 414 goes up to around 800/850 Hz at 6 dB/oct then the TAD TD2001 covers all the way upper.
Regarding the cross-over freqency, I don't have any problem to set it between 800 Hz and 1000Hz. Plus, you are right, there is a lot more choice in that range in the intention of a two ways design. I think that I stick on the cross-over suggested by Jean Hiraga in L'Audiophile magazine back in the days.
When you talk about cab's fine tuning, you mean that all of it would have to be redraw do you? On my side, I was looking for a project with the plan already available.
Regarding GPA, I couldn't find any graph with frenquency response. If anyone could oriented me to get them, I'll be happy. I already send them an email yesterday but no answer yet...
Finally, there are few things in your post that I don't understand. What do WRT, WG and expo horn mean?
Thanks again,
Sébastien
GM, your post is of great interest and contains a lot of informations. Meanwhile, I had the chance to talk to a member on Elektor which build the petite Onken and he told me that in its design, the Altec 414 goes up to around 800/850 Hz at 6 dB/oct then the TAD TD2001 covers all the way upper.
Regarding the cross-over freqency, I don't have any problem to set it between 800 Hz and 1000Hz. Plus, you are right, there is a lot more choice in that range in the intention of a two ways design. I think that I stick on the cross-over suggested by Jean Hiraga in L'Audiophile magazine back in the days.
When you talk about cab's fine tuning, you mean that all of it would have to be redraw do you? On my side, I was looking for a project with the plan already available.
Regarding GPA, I couldn't find any graph with frenquency response. If anyone could oriented me to get them, I'll be happy. I already send them an email yesterday but no answer yet...
Finally, there are few things in your post that I don't understand. What do WRT, WG and expo horn mean?
Thanks again,
Sébastien
Last edited:
WRT = with regard to
WG = wave guide
expo horn = exponential horn (I think)
I run full sized Onkens and am very pleased with their overall performance. I would recommend you get the new GPA 414-8B for all the same reasons GM mentioned.
I would cross closer to 1kHz than 800 were I you..
The Petite will be a lot of fun to build and listen to.
OT: Son et Image was interesting.
WG = wave guide
expo horn = exponential horn (I think)
I run full sized Onkens and am very pleased with their overall performance. I would recommend you get the new GPA 414-8B for all the same reasons GM mentioned.
I would cross closer to 1kHz than 800 were I you..
The Petite will be a lot of fun to build and listen to.
OT: Son et Image was interesting.
Greets!
You're welcome!
What XO did Hiraga recommend?
Well, the cab could be redesigned, but not a good plan unless you use actual measured specs of the drivers you buy; otherwise in the case of a fixed vent design such as the Onken, one normally will either use some form of EQ and/or stuffing density to smooth it out and/or block off one or more of the vents to tune it lower.
Haven’t seen a GPA response plot for the 414 and no longer have any originals, though the 802-8G SII has one in its specs pdf: http://www.greatplainsaudio.com/downloads/802-8G Spec Sheet.pdf
I see Kevin has already defined my [audio] 'net lingo’ that I’m notorious for.
For the non-audio related ones such as 'WRT', these are all listed here: NetLingo The Internet Dictionary
GM
You can also had a little 5" in OB like the good maid one by EMS speaker (they also without electric magnet) between the Onken and the horn for a faster and more realistic mid-bass to mid-tweet (let say around 200 to 1200 Hz) but maybe more a little more with the High pass because the OB and the small area.... but these one at EMS Speaker was maid if I remember for mid OB use with a very low QT < 0.20.
The plywood is important for the cabinet of the Onken, maybe some fellows at Melaudia.net could help you, they maid many Onken.
Last edited:
Hi Kevin,
Thanks for the explanations regarding the terms. I was not used to them.
It appears that the GPA 414 should be the right way to go. Only, I'm not equiped to make some mesures and I'm more interested to build the cabs and the cross-over but to follow a design already tried. This said, the Petite Onken seems to be a really nice speakers and I'd really like to move to a compression driver for mids and highs.
Unfortunately, it's the first time in 3-4 years that I choose not to go to the SSI. I went to the Montreal's DIY day in the beginning of March. That was it for this month.
I presented there my AMA 3x100 speakers. You can see them there:
AMA 3x100 (3 voies, 100 dB) - Page 14
Have a good day,
Sébastien
Hi GM, hi everyone,
Finally, I realized that the cross-over frequency of 600-650 Hz I had in my mind was more from Francis Ibre's Bien entendu, itinéraire d'un audiophile book than in Hiraga's article in L'Audiophile. My mistake. Ibre suggest that with such a design in two ways with a 12" woofer, the cross-over to the second way is aroud those frequencies. He says after that it's hard to find a good compression driver that could do the job. Hence the TAD TD2001.
After my reading, I realized that the member on Elektor's forum with the Petite Onken with TD2001 have a cross-over set at around 800-850 Hz, inspired by Hiraga's one applied to the VOT as described in Sound Pratices magazine number 11.
Have a good day,
Sébastien
Bonjour Eldam,
We have an excellent plywood supplied here in Quebec which I've build my AMA 3x100 with. It's the russian birch plywood grade "BB". I'm thinking about this one for the Petite Onken.
Regarding Melaudia, someone already talked to me about this forum and I think that I will post there soon.
Merci,
Sébastien
Excellent build quality on those speakers, and I enjoyed reading about your sonic impressions. With 100dB efficiency they should pair nicely with the future 300B PSE you mentioned.
I think you can build the Petite Onken design as published, I know of at least one person who has with great results. (Another friend is just getting going on a pair)
Remember Onkens are sensitive to driver qts which is in turn sensitive to dcr in the xo coils and amplifier source impedance. I started with a coil with slightly too much dcr and ended up replacing it with one with a slightly lower DCR which seemed to tighten up the bass a bit. All I am trying to say is be ready to tune things a little to the amplifier you are going to use.
OT: The show was pretty good, no SE in evidence (except a good sounding Elekit from Japan), but plenty of turntables, and tube amps (VTL, Manley, Quad, McIntosh mostly) more people using laptops and J River for media serving and much less iTunes than last year. I heard some good sound, and a lot of not so good sound.. lol
I think you can build the Petite Onken design as published, I know of at least one person who has with great results. (Another friend is just getting going on a pair)
Remember Onkens are sensitive to driver qts which is in turn sensitive to dcr in the xo coils and amplifier source impedance. I started with a coil with slightly too much dcr and ended up replacing it with one with a slightly lower DCR which seemed to tighten up the bass a bit. All I am trying to say is be ready to tune things a little to the amplifier you are going to use.
OT: The show was pretty good, no SE in evidence (except a good sounding Elekit from Japan), but plenty of turntables, and tube amps (VTL, Manley, Quad, McIntosh mostly) more people using laptops and J River for media serving and much less iTunes than last year. I heard some good sound, and a lot of not so good sound.. lol
its not been a year youv build your ama 3x100, why do you want to change speaker so fast?
Hi Kevin,
Thanks for the explanations regarding the terms. I was not used to them.
It appears that the GPA 414 should be the right way to go. Only, I'm not equiped to make some mesures and I'm more interested to build the cabs and the cross-over but to follow a design already tried. This said, the Petite Onken seems to be a really nice speakers and I'd really like to move to a compression driver for mids and highs.
Unfortunately, it's the first time in 3-4 years that I choose not to go to the SSI. I went to the Montreal's DIY day in the beginning of March. That was it for this month.
I presented there my AMA 3x100 speakers. You can see them there:
AMA 3x100 (3 voies, 100 dB) - Page 14
Have a good day,
Sébastien
Hi murphy,
I already asked myself a similar question. First, it's a pleasure for me to work with wood and solder. I like it. Plus, I'll have plenty of time this summer to work on such a project. So, first of all it's the fun of it.
Second, the Petite Onken came from a legacy which interest my greatly. Also, users report that this speaker sound nice. I also like the idea to try a compression driver for mids and highs. On my AMA 3x100, the mid is an "old" Audax PR170-ZO pretty hard to find nowadays.
Third, for sure I will compare it to my actual AMA 3x100 which I like very much. But even if the AMA are a 3 ways at 100 dB, the impedance goes down to 4 ohms. The Onken should be less sensitive but the impedance should be better. So I could see if I prefer one of the two.
Good evening,
Sébastien
Hi Kevin,
I'm glad to receive your positive comments regarding the AMA 3x100.
About the person you know who build the Petite Onken, could you tell me more? Does he use the GPA 414 woofer? What does he have above? Is it a 2 or 3 ways? Do you know the cross-over point?
Thanks,
Sébastien
OT: thanks for your quick report on SSI!
Hi GM,
Could you give me more detail about how you calculated this and get this result. I took Hiraga's formula from L'Audiophile's text. I reproduced it here (sorry, I'm not use to write formula with a computer):
Vo = 3,55 x 10 at5 x a at 4 x (l / k) at2 (4)
Mo fo at2
To discuss,
Sébastien
Greets!
OK, using Excel simple notation and correcting his formula to get the result in [L]:
Vo [L] = (((3.55*10^5*a^4)/(Mo*fo^2))*(1/1.57)^2/1000
Where:
a [cm] = driver effective radius, usually derived from its [Sd] spec
Mo [g] = driver moving mass with air load [Mms]
fo [Hz] = driver Fs
I have no idea what the (4) out at the end is for unless he meant ‘formula #4’.
Anyway, using his specs you should get ~155.85 L, though no clue why he rounded it down to 155 L since normally one rounds it up; then it’s just a matter of using another driver’s specs to calculate ‘Mo’ and ‘a’ if not provided.
Some folks use the long formulas to do this, though many now rely on a speaker design program to calculate it such as the WinISD Pro freeware I used. If you measure your drivers, then normally the program will automatically calculate it.
GM
OK, using Excel simple notation and correcting his formula to get the result in [L]:
Vo [L] = (((3.55*10^5*a^4)/(Mo*fo^2))*(1/1.57)^2/1000
Where:
a [cm] = driver effective radius, usually derived from its [Sd] spec
Mo [g] = driver moving mass with air load [Mms]
fo [Hz] = driver Fs
I have no idea what the (4) out at the end is for unless he meant ‘formula #4’.
Anyway, using his specs you should get ~155.85 L, though no clue why he rounded it down to 155 L since normally one rounds it up; then it’s just a matter of using another driver’s specs to calculate ‘Mo’ and ‘a’ if not provided.
Some folks use the long formulas to do this, though many now rely on a speaker design program to calculate it such as the WinISD Pro freeware I used. If you measure your drivers, then normally the program will automatically calculate it.
GM
Moi aussi , j'ai constaté cela , mais tant que le monde n'ont pas fait une comparaison cote à cote ils ne savent pas de quoi on parle . Moi je le descend jusqu'à 100Hz et laisse les deux premières octaves aux woofer .
Sure,
We know from 20 years that a two way with CD + Big mid-bass do not work most of the time...99,9 % it doesn't work. With good bass driver like the TAD 1601 it's easier but not ideal.
Sure than the Great Plain audio should be a very good unit for an Onken... Alnico if not electro magnet speakers is the way to go ( i have to win at lotery before building my own like that !). For what the OP need I would look at what Hiraga do with tuned semi-Onken with its last commercial design with an another famous driver (nearer of the OP's need!)
Not a specialist but having read all the Audiophile from Hiraga, my understanding is:
- low end has to be tuned : long and hard task : high order slope are always worst than slow one like 12 DB or less... vented are problematic because that !
- With CD : we need someting to match the heavy of the low with the light and fast of a CD (finally the same as planar...). Here 12 or 15" are too big (move too much air) and too heavy (for fastsness needs of mid... in relation to the needs of high mid for crossing with the CD)
10" could be ideal for mid to match with the low and the CD, but light one are paper cone : they loose details... some are good because transcient. Ideal should be like the one i talk about above (low QTS with OB is not a problem if no distorsion and no EQ)... here with 5" in OB I believe because the lake of air with the little 5" in the low end : we need :
- not a too low XO (200 to 300 HZ)
- because of that: 2 octaves of superimpose filter : so first order (which is hard or near impossible). the bass driver has to help here the littliest for moving air !
- of course electronic filter to set up XO of both drivers.
But ideal should be of course very light 8" to 10" OB : beginning at 100 to 150 hz: which is enough for directivity patern to match at around 1200 hz with a cd driver and horn (if not 4 ways design i mean). Electronic EQ is a problem because distorsion and transcient...(which a four ways design could avoid)
That's why 2 of my speakers use closed band pass design (Kef 104/2 ref and boston Lynfield 400L) : in highs: level slope is mechanicly first order and in lows: slope is smooth (12 DB slope) : mids beginn at 150 for the kef and 125 hz (second order elctrical but 90 hz first order mechanicly) for the Boston !
Show me a driver which is not paper, but light, with alnico magnet which work in OB between 150 hz and around 1000 hz and I will be happy... that's why EMS speakers do very smart drivers... Lavardin father maid high efficienty drivers for more 40 years...
We know from 20 years that a two way with CD + Big mid-bass do not work most of the time...99,9 % it doesn't work. With good bass driver like the TAD 1601 it's easier but not ideal.
Sure than the Great Plain audio should be a very good unit for an Onken... Alnico if not electro magnet speakers is the way to go ( i have to win at lotery before building my own like that !). For what the OP need I would look at what Hiraga do with tuned semi-Onken with its last commercial design with an another famous driver (nearer of the OP's need!)
Not a specialist but having read all the Audiophile from Hiraga, my understanding is:
- low end has to be tuned : long and hard task : high order slope are always worst than slow one like 12 DB or less... vented are problematic because that !
- With CD : we need someting to match the heavy of the low with the light and fast of a CD (finally the same as planar...). Here 12 or 15" are too big (move too much air) and too heavy (for fastsness needs of mid... in relation to the needs of high mid for crossing with the CD)
10" could be ideal for mid to match with the low and the CD, but light one are paper cone : they loose details... some are good because transcient. Ideal should be like the one i talk about above (low QTS with OB is not a problem if no distorsion and no EQ)... here with 5" in OB I believe because the lake of air with the little 5" in the low end : we need :
- not a too low XO (200 to 300 HZ)
- because of that: 2 octaves of superimpose filter : so first order (which is hard or near impossible). the bass driver has to help here the littliest for moving air !
- of course electronic filter to set up XO of both drivers.
But ideal should be of course very light 8" to 10" OB : beginning at 100 to 150 hz: which is enough for directivity patern to match at around 1200 hz with a cd driver and horn (if not 4 ways design i mean). Electronic EQ is a problem because distorsion and transcient...(which a four ways design could avoid)
That's why 2 of my speakers use closed band pass design (Kef 104/2 ref and boston Lynfield 400L) : in highs: level slope is mechanicly first order and in lows: slope is smooth (12 DB slope) : mids beginn at 150 for the kef and 125 hz (second order elctrical but 90 hz first order mechanicly) for the Boston !
Show me a driver which is not paper, but light, with alnico magnet which work in OB between 150 hz and around 1000 hz and I will be happy... that's why EMS speakers do very smart drivers... Lavardin father maid high efficienty drivers for more 40 years...
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.