My Next Horns

[Upfront]

Tuyen ill drop off my 300b monos and my pioneer Spec2 (and that cd player!) for you to have a play with when I'm back from Sydney. Be interesting to hear your thoughts on them.

[Paul Spencer]

Hi Tuyen,

Interesting post! Ok, now I've got some parameters, I did a quick model, and added the results to the blog:

http://redspade-audi...s-box-sims.html

In a nutshell, the box is far too big. The Goto woofer is very unique, with an abnormally high BL, in other words, a very strong motor and very high sensitivity. A BL of 37 and a Qts of 0.15 is almost unheard of!

Here is your box compared to a much smaller vented one, and a sealed box of the same size.

White = 360L tuned to 35 Hz ... I'm guessing the tuning

Red = 60L tuned to 40 Hz

Blue = sealed 360L

You definitely don't want your response to look like that.

First thing to try is to start stuffing the ports. You will transition from what you have now, to the sealed box. You might find an in-between that you like.

Notice the drooping response in red? It's chosen as a rough guess that your room has some serious peaks that will actually bring up the apparent bass level. This is pretty hit and miss, but with a semi blind guess I'd be looking at a vented box that looks more like that.

I suspect your Goto woofer is actually begging for a horn with all that BL.

It's a bit of a shame you gave up on measurements, but it takes some persistence. A lot of people get a bit lost in the maze when they find that things don't work as they expect they should. You get something to measure "good" and it doesn't sound good! It just means you are at the start of the learning curve.

With horns, I've heard all kinds, ranging from some pretty bad ones and some very nice ones as well. Horns can fail in a spectacular way, they can also succeed in the same way. They can also be both brilliant and nasty.

In some cases, you can look at a horn and see problems, more so when you know a few details about it. If someone says "this is an 80 Hz bass horn with driver XYZ and it runs up to 2k with an expo profile" ... I might look at it and be able to say the path length is far too short, or the mouth far too big. Sometimes horns are out by so much that the mouth is 10 times bigger than it should be, because some people get caught up in looking like another horn they have seen online, but they haven't modelled it, understood the basics of how it works or measured it. So without hearing a horn, knowing a few details can reveal basic design flaws. With horns, some things are quite finely tuned, others you can get away with being a little bit out.

Designing horns is a matter of balance, with many of the parameters needing to "talk" to each other. The size of the mouth, the path length of the horn, the profile chosen, the size of the front and rear chambers, the throat area - all these things interact and to model a horn, the first design step, you have to get them to all work together.

Horns are the most difficult to get right of all speakers. They have more ways they can go wrong. The have more parameters that interact with each other. There is so much more that you have to understand to design a well sorted horn. If you can design a horn speaker from scratch, get all the parts to work together as a whole, understand each parameter and how it relates to all the others, work through all the little decisions understanding the compromises you are making, and come out the other end with a speaker that is well-rounded, well sorted and not a "one trick pony" ... if you can do all that then you've not only worked very hard for the smile on your face when you fire them up, but now the sky is the limit and you can design any speaker you want.

In reality, most horn builders stop short and either give up on horns or they are fairly tolerant of certain flaws.

Back to your situation ... here's what you could do if you want some bass horn fun. Build a pair of bass horns the run across your entire front wall, each mouth firing to its respective corner. You would have to fold it a little and it would be like stretching your existing speakers to cover the space in between. You'd have to figure out your rack arrangements, but I'd also put your other drivers on a baffle, and toe them in, trying toe in far enough that the axes cross in front of your listening position, paying attention to the impact on imaging in particular. Also try listening to either side of your listening position when you do this, noticing how it affects the sweet spot. A good bass horn with some extension needs about twice the volume of your existing boxes, or a more suitable vented box needs to be more like 1/6 the volume. The very high motor strength of your drivers that makes them a good candidate for horn loading, also means they want a small box when ported.

[Steve M]

Have recently heard Tuyen's Horns, very nice set-up especially for the dynamic of horn-effect, strongly projected sound, natural acoustic.

To my ears, just needs a bit of bass depth for texture all round ...soon to be resolved I assume due to purchase of Atilsley's horn loaded bass boxes.

Btw, Tuyen is a really nice 'tuned-in' guy sans bs from my experience ...

Cheers,

Steve.

[THOMO]

"but 'horny' friends who do use them, say it's the final frontier?"

The potential is there with bass horns to get very clean bass, with massive headroom and no real SPL limits. But nearly all bass horns are seriously compromised and that results in inferior time domain performance, and they can easily be less accurate than a good direct radiator. A lot of bass horn aren't properly designed, with the critical parameters correctly optimised. In fact there are so many horns out there that are messed up, that some are surprised at what a correctly designed one looks like.

"ha ha! To be honest, I don't know much about damping factor, only recently starting reading up about it to get some idea. I have been living with flabby bass for past few months thinking it was mainly due to the room! "

The worst case scenario with damping factor would be this. A vented box with a valve amp with ultra low damping factor, in a room with no damping at all - think cave. So what happens? Below the tuning of the vented box, the driver becomes unloaded - this occurs with all vented boxes. We are now counting on the suspension system of the driver and the damping factor of the amp to keep things under control. In this worst case scenario the driver flops about as if there is extreme LF rumble. The driver starts acting like a microphone, similar situation to feedback in a PA system screeching. You see the cone moving like crazy, big excursions. The bottom end becomes fuller, boosted and bloated. So you block the vent and the bass becomes tighter as the box is now controlling excursion, the problem is cut back a great deal. The problem even in this extreme case is as much pragmatic as it is sonic - the driver doesn't like that kind of excursion and the situation causes a fuller bass sound.

Boomy bass is likely to be caused by a cavernous room, concrete floor and ceiling and brick walls. Flabby bass suggests issues with integration and the box itself - have you got it well optimised? Your box is quite large and if the driver has a strong motor (high BL, low Qts) then you can get a peaky bottom end and early roll off. You may not have the right match between driver and box. At this point anything is possible - could be the room, the bass box itself or both.

Exactly what I was getting with some Tannoy HPD 385s.Low Qts ,low FS drivers in an all masonry room. Onken type boxes helped a bit but on some recordings there was just too much bass.Other people have described it as feeling a bit beaten up by the bass.People in the UK who typically have suspended timber floors over basements seem to have fewer issues with these drivers.Many also favour back loaded horns like the GRF rectangle that would give less low bass output and also have different damping.

Perhaps this is also why most pro type 15 inch drivers have higher resonant frequencies/ stiff surrounds/short excursions.The problem with them is often too little low bass output unless you use equalisation.

[Owen Y]

2A3PP monos? custom? tell us more!

With these spkrs (& earlier Mauhorns), I had built & used 300BSE & 2A3SE amps previously.

The present amps are 2A3PP, built from a well spoken-about Valves World Japanese cct - which provides a bit more power & importantly, lower output impedance / higher damping factor => better LF control.

Output is 10-12wpc

Tube line-up is 76 - 6A6 - 2A3PP - Partridge 6k6:8 OPT.

Preamp is a cross between the Berman & JEL, 76 - 6SN7 with Sowter TVC vol control. Sometimes referred to as the Korneff preamp.

(The pre on LHS above is an earlier OPT coupled 5687 job, which I eventually stripped out for the one below.) Cheers.

Edited September 8, 2012 by Owen Y

[davewantsmoore]

BL of 37 and a Qts of 0.15

I suspect your Goto woofer is actually begging for a horn with all that BL

0_o Wow strong. Concur re: horn

The claimed sensitivity of this driver (110db) is pretty insane (or am I reading that wrong from somewhere... i've only skim read the end of this thread --- too busy look at the pics) :-)

[Paul Spencer]

Dave, I have no idea what they claim, but for a 15" driver, 99db shown above is very high. Generally a 15" driver that is around 98 db will struggle to get to 50 Hz. A driver with a very strong motor and very low moving mass and very high sensitivity can only maintain it's sensitivity on the top end. You can get a 6" driver to get that high in sensitivity, but it rolls off below 600 Hz! If you filter this driver and shed some sensitivity, you can get to 40 Hz with more like 95 db. When a single number is given, it means nothing at all. The true sensitivity that matters, is what can be achieved at the lowest point of its useful bandwidth. Some quote the maximum or the average, but neither are meaningful. The laws of physics and the limits of dynamic drivers dictate that there is a maximum sensitivity for a given piston area and extension. If Goto quote their 15" as 110 db then they could be referring to horn loaded sensitivity. It's rare to find more than 100 db in a direct radiator, because it ends up not being useful in many designs, due to limited bandwidth.

[davewantsmoore]

Yes, I thought this was the 110db driver (which is insane) .... and my first thought was the same re: quoting it's horn loaded efficiency. I had a real quick search to try and confirm that and couldn't see any reference to that.

The laws of physics and the limits of dynamic drivers dictate that there is a maximum sensitivity for a given piston area and extension

Yes, I think lots of people miss this fact.... and also like you say, that a single number doesn't tell the story.

As you know I was confused about bass / midbass cabinets for a long time .... my prefs are now totally different after some experimenting. Group delay 'sealed' the deal (pardon the pun).

To get back on topic. I can only second the suggestion to try sealing the cabinets up some and lowering their volume.

... and also to get the measurement gear back. Don't bother with trying to flatten bumps in a completed system (like you've noted, its an OCD nightmare :-)) ..... use the measurement tools for "system design".

[tuyen]

Tuyen ill drop off my 300b monos and my pioneer Spec2 (and that cd player!) for you to have a play with when I'm back from Sydney. Be interesting to hear your thoughts on them.

Oh, thanks Linc! Would very much appreciate giving them a go. Considering we only live like 3 streets apart, just come over and listen for yourself mate Catch up when you get back from Sydney.

[tuyen]

Have recently heard Tuyen's Horns, very nice set-up especially for the dynamic of horn-effect, strongly projected sound, natural acoustic.

To my ears, just needs a bit of bass depth for texture all round ...soon to be resolved I assume due to purchase of Atilsley's horn loaded bass boxes.

Btw, Tuyen is a really nice 'tuned-in' guy sans bs from my experience ...

Cheers,

Steve.

Hi Steve,

Thanks! Your cheque is in the mail.

Anyway, didn't end up buying the tuba horn bass units, as there was local buyer who ended up picking them up from Andrew.

No to worry though, as I believe part of the issue was the 6v6gt power amp was not controlling the woofers very well. Have since did some experimentation and found that by hooking up the 20watt class a hiraga amp, the bass has tightened up with more extension, definition/texture, punch & control. Close to how it sounded when I was running my system actively with the Pioneer D-23 electronic crossover. So that experiment was a bit of an eye opener.

Should be getting back my accuphase based 50watt class AB amp today from Bevan, so will give that a go to see how that sounds as well. I'm current running my system in bi-amp mode (hiraga for woofers, 6v6gt amp for horn channels) and it seems to be working quite well so will try adhere to this style of configuration. Would love to try your Krell and F5 on the woofers if your offer is still up?

But as Paul has noted, the cabinets aren't optimised for the woofers, so may look at designing horn loaded cabs for the woofers! Now that'll be fun!

Did you end up pulling your edgar horn bits and pieces out of the attic yet?

[THOMO]

Dave, I have no idea what they claim, but for a 15" driver, 99db shown above is very high. Generally a 15" driver that is around 98 db will struggle to get to 50 Hz. A driver with a very strong motor and very low moving mass and very high sensitivity can only maintain it's sensitivity on the top end. You can get a 6" driver to get that high in sensitivity, but it rolls off below 600 Hz! If you filter this driver and shed some sensitivity, you can get to 40 Hz with more like 95 db. When a single number is given, it means nothing at all. The true sensitivity that matters, is what can be achieved at the lowest point of its useful bandwidth. Some quote the maximum or the average, but neither are meaningful. The laws of physics and the limits of dynamic drivers dictate that there is a maximum sensitivity for a given piston area and extension. If Goto quote their 15" as 110 db then they could be referring to horn loaded sensitivity. It's rare to find more than 100 db in a direct radiator, because it ends up not being useful in many designs, due to limited bandwidth.

Pretty much what I found when I was experimenting with high sensitivity low low Qts pro type drivers.You can't get much low bass out of them at low power input.Very few companies publish frequency responses for these drivers at 1 watt input which in a domestic setting is about what you would use.RCF did and these responses showed very little bass below 100hz at 1 watt.Then if you go with lower sensitivity drivers they do not seem to integrate properly with horn midranges[they sound slow].

I am sure there are some exceptions but I am yet to hear one.The medium sensitivity high Qts drivers like some of the cheap Eminence drivers do not seem to sound quite right either.

The Fostex 15 inch drivers might work.I have heard them sounding very good in a big vented box .

Using Tannoy dual concentics just for bass is a possibility.They are more medium to high sensitivity,do low bass at low power input, and are designed to reproduce low midrange.

Edited September 8, 2012 by THOMO

[statman]

I've never taken the damping factor specification of an amp into consideration, but now I want to figure out what is the optimal value to drive the woofer, it seems like there is no clear answer?

Tuyen

This is one of my pet subjects. Very few people do take DF into consideration when choosing amps, and if people realized how much variation of frequency response is caused by the DF of an amp into a speaker load they would be having another look at their prized SE amps. Have a look at the chart at http://www.supratekaudio.blogspot.com.au/2011/06/circuit-design.html This chart shows response of very good SE amp into a speaker load, the black line is response into the speaker, the other lines are resistive loads. +/- 2 dB variation is actually pretty good for a SE amp, a 300B would be a lot worse. We can easily hear a 0.5dB variation, so imagine what a 2dB variation does to the response, which is why some call SE amps "tone controls".

A better performing SE amp will use lower plate resistance tubes, or change the ratios in output transformer to increase damping factor, but a SE amp is only really legitimate, imo, when used with a speaker that has a ruler flat impedance, and there are very few of them, maybe a horn with a good driver and used from lower midrange (say 300Hz) might be reasonably accurate.

Bass definitely needs a better DF to give deep tight bass. I think PaulS is right in suggesting your speaker box has a lot more to do with your flabby bass than the amp, but as you discovered the DF is important factor in good bass (and everything else)

I dont think your Hiraga SS amp would have a particulary good DF, and it does seem a real shame to have a nice active horn system such as yours and not use tube amps.

Owen is on the right track with PP tube amps and he's using a transformer ratio a bit higher than normal with his 2A3 amps to give a better DF.

Something like the 2A3 which is a lowish plate resistance tube will have a much better DF than a higher power amp using 211 or 845, which are high plate resistance tubes. 845 amps have big "boomy" bass -a reflection of the DF.

A little negative feedback helps a lot with DF , you just have to be careful not to use too much and affect the phase performance which screws up the soundstaging, imaging and dynamics, but there is a happy medium where DF is considerably improved and sonics are unaffected. Stiff power supplies help a lot, shunt regulation is helpful in avoiding big electrolytic power supply capacitance, which can cause problems in non bass area freqs.

Positive feedback is coming back into fashion, I've been using it for nearly 10 years and its the only way I can get my active electrostatic/planar magnetic ribbon/moving coil bass driver hybrid speakers to work with tube amps. Positive feedback doesn't have the same degree of degradation that negative feedback has when used excessively and I can't understand why it isn't more widely used.

Even the 300B when used with a bit of negative feedback, a bit of positive feedback and used in PP circuit will make your GOTO bass drivers growl with deep tight bass, in the right enclosures of course.

Sure the big Krells and other SS power amps will probably go even deeper, but come on, Goto drivers and SS amps? What would Mr GOTO say?

I love what your doing, haven't used horns myself for a long time as I got into an accuracy thing , but looking at your pictures brought back memories of fun times- even ordered a pair of Paul's S3 to use with some tube amps sitting in the shed.

Owen, I haven't looked at the Valves World site for a few years, a few new designs there, but still all from the Sound Practices era, some interesting Shisido type positive grid Class A amps (with very low DF) 211,845 amps with HT voltage under 1KV (why)

but I did like the 212E SE amp driven by a 211 AND an 845 connected in parallel, each with their own cathode bias resistor and both outputs driving the 212E. The 211 and 845 sound so different, wonder what they would be like in tandem, very interesting.

[Steve M]

Hello Tuyen,

No problems, come and grab the Krell KSA50 and First Watt F5 amp, the Krell is a good workhorse and very good on stats and ribbons while the F5 is probably more finessed.

No, haven't dragged the Edgar horns out yet, still enjoying the Nak spkrs. Have got some Accuton midranges coming from the Netherlands in a week, so when they arrive I will set up my 80L Eton bass reflex boxes and DBX xo which will allow me to try both a three way Accuton Raven R2 spkr or with Edgar module on top.

With respect to your bass and Onken boxes, the things Paul Spencer said are true, ie, if the box is not properly sized in terms of TS Parameters, you will get a boomy hollowy sound. With my Onken box the back panel slid in and out to change box volume for tuning and I also stuffed some of the ported slots to experiment. The WIN-ISD speaker modeling program is your friend and guide here.

You can also put some cheap basketballs or other plastic sports balls or bricks (too much lifting and laying though) into the Onken boxes to reduce volume.

Have fun ...

Steve.

[Paul Spencer]

Pretty much what I found when I was experimenting with high sensitivity low low Qts pro type drivers.You can't get much low bass out of them at low power input.Very few companies publish frequency responses for these drivers at 1 watt input which in a domestic setting is about what you would use.RCF did and these responses showed very little bass below 100hz at 1 watt.Then if you go with lower sensitivity drivers they do not seem to integrate properly with horn midranges[they sound slow].

I am sure there are some exceptions but I am yet to hear one.The medium sensitivity high Qts drivers like some of the cheap Eminence drivers do not seem to sound quite right either.

The Fostex 15 inch drivers might work.I have heard them sounding very good in a big vented box .

Using Tannoy dual concentics just for bass is a possibility.They are more medium to high sensitivity,do low bass at low power input, and are designed to reproduce low midrange.

Thomo,

I have to wonder what drivers you are looking at, that don't come with frequency response with 1w input. Must be some boutique, exotic drivers? There are few drivers I'd consider using without that information. With dedicated bass drivers I want to see if there is early top end roll-off due to inductance - you can't tell just from voice coil inductance alone. I've measured one driver that was 6 dB down at 300 Hz! I also want to see how far I am from any cone breakup. I also like to see how the sensitivity is achieved, if it has an ultra high sensitivity due to a rising response. Without this information I could be dealing with a useless driver.

With respect to sensitivity and perceived bass speed, be careful of the connections you make. Sensitivity is just a way to arrive with a different fuel economy. Obviously higher sensitivity lets you use valve amps, which then bring in their own traits. The real issue is not so much the sensitivity, as if we could isolate that alone, we would no doubt find the audible differences aren't what we might expect. All else being equal, higher sensitivity would tend to preserve dynamics due to minimising thermal and mechanical effects which can rob dynamics. However, in reality, these effects are IMHO swamped by other side effects of any chosen design.

You can get very good bass in many different ways. You can use a very low sensitivity high quality subwoofer with only 85 db 1w1m, or you can use a 95 dB pro woofer. If you do it right, you can get a very good result either way. However, in the vast majority of cases, it has not been done in the way that I would call "right." My idea of doing it properly starts with designing the box and driver combination, powering it correctly, optimising it in the room with measurements and treating the room seriously until you meet bass damping targets.

If you can only get the kind of bass you want with very high sensitivity, then the chances are what you ear hearing and appreciating is a side effect, and not directly related. The first suspect would be frequency response - you probably have a particular curve you are getting that sounds right to you. It just might be that the higher sensitivity drivers allow some room modes to fill in the bottom end.

I currently have my room set up with some 94 db 10" vented speakers, no subs, no EQ - they are just sitting there, they roll off at 50 Hz anechoic, and my room has a big 44 Hz peak. It works pretty well as a plug and play arrangement, but it's not nearly as good as when I have the subs running and do it more seriously.

I now suspect I have more idea what is behind the crappy hollow bass of quite a few vintage bass cabs. Wrong driver and box combination, out by a country mile. There are also some drivers floating around that were made before the concept of TS Parameters was conceived.

[Steve M]

Mick/statman,

Interesting stuff from you as usual, what I'm wondering is in terms of the amplifier for the bass, is damping factor even an issue once you get any competent solid state amp into that position. What I mean is most good ss amps (especially ones that double their power into halving the impedence), all seem to have good damping characteristics?

Steve.

[statman]

Mick/statman,

Interesting stuff from you as usual, what I'm wondering is in terms of the amplifier for the bass, is damping factor even an issue once you get any competent solid state amp into that position. What I mean is most good ss amps (especially ones that double their power into halving the impedence), all seem to have good damping characteristics?

Steve.

Pretty much true, I do tend to use SS amps on bass <200Hz myself, but with a fully fledged horn system its kind of sacrilegous dont you think? Its not that hard to build a good tube amp that has a reasonable DF.

[Owen Y]

You can also put some cheap basketballs or other plastic sports balls or bricks (too much lifting and laying though) into the Onken boxes to reduce volume

Bruce Edgar used plastic water bottles when determining chamber volume reductions in his horns - easy to just count up the 2-litre bottles.

[NIKO]

Goto drivers and SS amps? What would Mr GOTO say?

Hello

Going by reports Mr Goto like their Bakoon active system .

Chatting with 56 and the bass valve amp circuit Lucas is drawing up for him Output transformer need to be able to cover lower frequency .The output transformer for Mal has been measured 0.9db @ 10hz .

So don't forget about the quality of the output transformer .

yassou

[tuyen]

Hi Tuyen,

Interesting post! Ok, now I've got some parameters, I did a quick model, and added the results to the blog:

http://redspade-audi...s-box-sims.html

In a nutshell, the box is far too big. The Goto woofer is very unique, with an abnormally high BL, in other words, a very strong motor and very high sensitivity. A BL of 37 and a Qts of 0.15 is almost unheard of!

Here is your box compared to a much smaller vented one, and a sealed box of the same size.

White = 360L tuned to 35 Hz ... I'm guessing the tuning

Red = 60L tuned to 40 Hz

Blue = sealed 360L

You definitely don't want your response to look like that.

First thing to try is to start stuffing the ports. You will transition from what you have now, to the sealed box. You might find an in-between that you like.

Notice the drooping response in red? It's chosen as a rough guess that your room has some serious peaks that will actually bring up the apparent bass level. This is pretty hit and miss, but with a semi blind guess I'd be looking at a vented box that looks more like that.

I suspect your Goto woofer is actually begging for a horn with all that BL.

It's a bit of a shame you gave up on measurements, but it takes some persistence. A lot of people get a bit lost in the maze when they find that things don't work as they expect they should. You get something to measure "good" and it doesn't sound good! It just means you are at the start of the learning curve.

With horns, I've heard all kinds, ranging from some pretty bad ones and some very nice ones as well. Horns can fail in a spectacular way, they can also succeed in the same way. They can also be both brilliant and nasty.

In some cases, you can look at a horn and see problems, more so when you know a few details about it. If someone says "this is an 80 Hz bass horn with driver XYZ and it runs up to 2k with an expo profile" ... I might look at it and be able to say the path length is far too short, or the mouth far too big. Sometimes horns are out by so much that the mouth is 10 times bigger than it should be, because some people get caught up in looking like another horn they have seen online, but they haven't modelled it, understood the basics of how it works or measured it. So without hearing a horn, knowing a few details can reveal basic design flaws. With horns, some things are quite finely tuned, others you can get away with being a little bit out.

Designing horns is a matter of balance, with many of the parameters needing to "talk" to each other. The size of the mouth, the path length of the horn, the profile chosen, the size of the front and rear chambers, the throat area - all these things interact and to model a horn, the first design step, you have to get them to all work together.

Horns are the most difficult to get right of all speakers. They have more ways they can go wrong. The have more parameters that interact with each other. There is so much more that you have to understand to design a well sorted horn. If you can design a horn speaker from scratch, get all the parts to work together as a whole, understand each parameter and how it relates to all the others, work through all the little decisions understanding the compromises you are making, and come out the other end with a speaker that is well-rounded, well sorted and not a "one trick pony" ... if you can do all that then you've not only worked very hard for the smile on your face when you fire them up, but now the sky is the limit and you can design any speaker you want.

In reality, most horn builders stop short and either give up on horns or they are fairly tolerant of certain flaws.

Back to your situation ... here's what you could do if you want some bass horn fun. Build a pair of bass horns the run across your entire front wall, each mouth firing to its respective corner. You would have to fold it a little and it would be like stretching your existing speakers to cover the space in between. You'd have to figure out your rack arrangements, but I'd also put your other drivers on a baffle, and toe them in, trying toe in far enough that the axes cross in front of your listening position, paying attention to the impact on imaging in particular. Also try listening to either side of your listening position when you do this, noticing how it affects the sweet spot. A good bass horn with some extension needs about twice the volume of your existing boxes, or a more suitable vented box needs to be more like 1/6 the volume. The very high motor strength of your drivers that makes them a good candidate for horn loading, also means they want a small box when ported.

Great informative post as usual, Paul. Thank you! You are right about the cabs being too big for the drivers, as over the many photos of goto horn setups I have come across, the cabs used are much smaller than the 360L onkens I'm using. You are also correct regarding the drivers begging to be horn loaded. As one of the 'ultimate' sounding Goto Unit systems in Europe (Klaus' in Germany), is in fact running a pair of similar drivers with a front loaded horn. I have been recommended to do the same with my drivers by many. Only problem is, the size of the horn they recommended is as big as my room (photo below).

And now here are some photos of multiway goto setups which are using goto woofers, taking note of the woofer cab size:

http://www.audionaut...ce_2007_eng.htm

Come to think of it, I have a some 125L sealed cabs that the AE woofers were in, that I can use to put the goto woofers into. Would they be worth a shot? Porting them is possible to, just that not sure what size port is required... any chance you could please advise, Paul?

Photo of the cab compared the onken below:

Regarding measurements, I didn't sell off all my equipment Still got my jaycar spl meter + tone generator software! hah

Had Bevan pop over on the weekend with his tools, and managed to do some measurements.. will post findings soon.

[Owen Y]

As one of the 'ultimate' sounding Goto Unit systems in Europe (Klaus' in Germany), is in fact running a pair of similar drivers with a front loaded horn. I have been recommended to do the same with my drivers by many. Only problem is, the size of the horn they recommended is as big as my room (photo below).

Hi tuyen,

Those look to be a small driver & the horns are almost 'free space' horns (with v little floor loading).

The following v small horn box (by my pal Johari in S'pore) was used with a 15" driver....

Small mouth size maybe, but it is floorloaded/coupled.

The bigger the driver, the shorter the horn.

The mouth size requirement however does not change for any chosen design frequency.

However, the way you locate & 'load' the horn mouth is significant...

Corner placement will reduce mouth size by 1/4 (compared with floor loading).

If you want to go down this route, I would be looking at corner horns, with your mid/HF drivers located inboard.

As said, your masonry walls/floor will be a great advantage

[Owen Y]

Something like the Klipschorn or Tannoy corner designs maybe.

Or, something more adventurous like a 'floor-horn'... with your mid/HF transducers located on top....

(It should be able to be made more compact.)

[Paul Spencer]

Tuyen, 125L is a bit on the large size also, but workable. You'll need to tune around 35 Hz, which means 44mm length for a 100mm piece of PVC pipe. Or if the box is 36mm thick at any point, cut a single 80mm circle. Or if the box is just 18mm thick, then cut a 7cm circle.

[tuyen]

Thanks for all your posts so far guys. Much appreciate it!

Hey Paul, I think the cabinets might be pretty thick. Possibly 25-36mm. I'll have to confirm with Bevan as I had got them from him. But basically, I would just need to cut a simple hole in the cab? No need any special tuning pipes? Or do I still need the tuning pipe no matter what size the thickness is?

Anyways, here are some measurements we got:

The 3 upper horn channels were measured nearfield to get that reading for this one. Don't have an overall freq response of system at listening spot though.

The red line shows measured response of woofer nearfield, purple at about 1.5m away, and orange at listening position. So looks like your sim was pretty accurate! big dip around 60hz... and hump at 40hz, hence contributing to the 'boominess'... I have much to learn from you!

Edited September 10, 2012 by tuyen

[tuyen]

With these spkrs (& earlier Mauhorns), I had built & used 300BSE & 2A3SE amps previously.

The present amps are 2A3PP, built from a well spoken-about Valves World Japanese cct - which provides a bit more power & importantly, lower output impedance / higher damping factor => better LF control.

Output is 10-12wpc

Tube line-up is 76 - 6A6 - 2A3PP - Partridge 6k6:8 OPT.

Preamp is a cross between the Berman & JEL, 76 - 6SN7 with Sowter TVC vol control. Sometimes referred to as the Korneff preamp.

(The pre on LHS above is an earlier OPT coupled 5687 job, which I eventually stripped out for the one below.) Cheers.

Very cool. Love reading about diy gear. The pre/power combo seems to be all built with performance and sound quality as top priority. Well done! My setup currently is still far from optimised.. one irritant for me is that every amp has a 'dirty' volume pot in the way. When I should just have 1 pot for the preamp and that's it.. long way to go! Thanks for sharing more info on your system.

Tuyen

This is one of my pet subjects. Very few people do take DF into consideration when choosing amps, and if people realized how much variation of frequency response is caused by the DF of an amp into a speaker load they would be having another look at their prized SE amps. Have a look at the chart at http://www.suprateka...uit-design.html This chart shows response of very good SE amp into a speaker load, the black line is response into the speaker, the other lines are resistive loads. +/- 2 dB variation is actually pretty good for a SE amp, a 300B would be a lot worse. We can easily hear a 0.5dB variation, so imagine what a 2dB variation does to the response, which is why some call SE amps "tone controls".

A better performing SE amp will use lower plate resistance tubes, or change the ratios in output transformer to increase damping factor, but a SE amp is only really legitimate, imo, when used with a speaker that has a ruler flat impedance, and there are very few of them, maybe a horn with a good driver and used from lower midrange (say 300Hz) might be reasonably accurate.

Bass definitely needs a better DF to give deep tight bass. I think PaulS is right in suggesting your speaker box has a lot more to do with your flabby bass than the amp, but as you discovered the DF is important factor in good bass (and everything else)

I dont think your Hiraga SS amp would have a particulary good DF, and it does seem a real shame to have a nice active horn system such as yours and not use tube amps.

Owen is on the right track with PP tube amps and he's using a transformer ratio a bit higher than normal with his 2A3 amps to give a better DF.

Something like the 2A3 which is a lowish plate resistance tube will have a much better DF than a higher power amp using 211 or 845, which are high plate resistance tubes. 845 amps have big "boomy" bass -a reflection of the DF.

A little negative feedback helps a lot with DF , you just have to be careful not to use too much and affect the phase performance which screws up the soundstaging, imaging and dynamics, but there is a happy medium where DF is considerably improved and sonics are unaffected. Stiff power supplies help a lot, shunt regulation is helpful in avoiding big electrolytic power supply capacitance, which can cause problems in non bass area freqs.

Positive feedback is coming back into fashion, I've been using it for nearly 10 years and its the only way I can get my active electrostatic/planar magnetic ribbon/moving coil bass driver hybrid speakers to work with tube amps. Positive feedback doesn't have the same degree of degradation that negative feedback has when used excessively and I can't understand why it isn't more widely used.

Even the 300B when used with a bit of negative feedback, a bit of positive feedback and used in PP circuit will make your GOTO bass drivers growl with deep tight bass, in the right enclosures of course.

Sure the big Krells and other SS power amps will probably go even deeper, but come on, Goto drivers and SS amps? What would Mr GOTO say?

I love what your doing, haven't used horns myself for a long time as I got into an accuracy thing , but looking at your pictures brought back memories of fun times- even ordered a pair of Paul's S3 to use with some tube amps sitting in the shed.

Owen, I haven't looked at the Valves World site for a few years, a few new designs there, but still all from the Sound Practices era, some interesting Shisido type positive grid Class A amps (with very low DF) 211,845 amps with HT voltage under 1KV (why)

but I did like the 212E SE amp driven by a 211 AND an 845 connected in parallel, each with their own cathode bias resistor and both outputs driving the 212E. The 211 and 845 sound so different, wonder what they would be like in tandem, very interesting.

Thanks Mick. Be interested to hear how you find the S3 kit when its all setup and running.

I think there are a few Goto setups that are using low powered class A ss amps actually. I know a few in europe that are running actively with multiple 8watt hiraga le monstre amps. Not sure how good they sound though. At the moment, I'm enjoying trying various tube power amps to run the horn channels and still a bit undecided what amps to run the woofers. I think I firstly need to sort out the actual woofer/cabinet combination before I spend more time looking at different amps (even though they can make big effect on the sound as well).

Hello Tuyen,

No problems, come and grab the Krell KSA50 and First Watt F5 amp, the Krell is a good workhorse and very good on stats and ribbons while the F5 is probably more finessed.

No, haven't dragged the Edgar horns out yet, still enjoying the Nak spkrs. Have got some Accuton midranges coming from the Netherlands in a week, so when they arrive I will set up my 80L Eton bass reflex boxes and DBX xo which will allow me to try both a three way Accuton Raven R2 spkr or with Edgar module on top.

With respect to your bass and Onken boxes, the things Paul Spencer said are true, ie, if the box is not properly sized in terms of TS Parameters, you will get a boomy hollowy sound. With my Onken box the back panel slid in and out to change box volume for tuning and I also stuffed some of the ported slots to experiment. The WIN-ISD speaker modeling program is your friend and guide here.

You can also put some cheap basketballs or other plastic sports balls or bricks (too much lifting and laying though) into the Onken boxes to reduce volume.

Have fun ...

Steve.

thanks for the advice Steve. I can't really hear the hollowness, but measurements do show that it is evident. I do hear the boominess issue that the sims show though. I'm just going to fit the woofers into the sealed cabs that the AE woofers are in, then also try porting them and see how we go. Can't be bothered experimenting fitting stuff into the onken as plan to eventually pair them with the better matching AE woofers and offload them to a friend.

Pretty much true, I do tend to use SS amps on bass <200Hz myself, but with a fully fledged horn system its kind of sacrilegous dont you think? Its not that hard to build a good tube amp that has a reasonable DF.

My system is hardly fully fledged horn though. Only using a horn midlow channel.. The woofers are standard 15" drivers, the midhigh channel is a beyma AMT ribbon styled driver and the tiny 3000hz horn on the goto tweeter is too small to be considered a horn I think I'll stick to using a ss power amp to power the woofers for now, atleast until I sort out the bass/cabinet combo..

Hi tuyen,

Those look to be a small driver & the horns are almost 'free space' horns (with v little floor loading).

The following v small horn box (by my pal Johari in S'pore) was used with a 15" driver....

Small mouth size maybe, but it is floorloaded/coupled.

The bigger the driver, the shorter the horn.

The mouth size requirement however does not change for any chosen design frequency.

However, the way you locate & 'load' the horn mouth is significant...

Corner placement will reduce mouth size by 1/4 (compared with floor loading).

If you want to go down this route, I would be looking at corner horns, with your mid/HF drivers located inboard.

As said, your masonry walls/floor will be a great advantage

The horn that the kid is standing in is using higher end 15" goto woofer that I am using mate. The other big horn next to it is using a compression bass driver, hence the smaller 6" horn throat size.

That small horn box Johari uses looks cool. does it play low? I know Stefano (twogoodears) is using something similar for his setup.

Something like the Klipschorn or Tannoy corner designs maybe.

Or, something more adventurous like a 'floor-horn'... with your mid/HF transducers located on top....

(It should be able to be made more compact.)

floor-horn out of the question! corner horns could be possible...

[Owen Y]

Am I reading to plots correctly?... Relative to 1khz, you could say that (i) the whole LF spectrum is up a bit, with a couple of peaks, at ~42hz & ~95hz.

Coincidentally, you have room width that looks ~4.0m.... ie. a 43hz width standing wave fundamental?

Try an experiment...place your spkrs (C/L front of cabinet) at 1/3rd or 1/5 width & depth dimension into the room.

eg. 0.8m from ea side wall (assuming 4.0m rm width) & say 1.0m from rear wall for say a 5.0m m length (adjust to suit.

Similarly also, your listening position should be away from rear wall by 1/3, 1/5, 2/7 rm length (ie. odd fractions).

That might give you more uniform LF response anyway. Cheers.