Zaph Audio SB12.3 or Troels' DTQWT?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
One of the issue for a novice is to translate a schematic of the xo to a physical design, but in this case Troels provides a figure of the xo, so you only have to follow that figure. It's really easy.

However I'd be worried about soldering. Look at some tutorials and practice on scrap components. Basically you need to be fast and make good joints. A solder with a small component of silver (2%) is of audio quality.

For me the most difficult work on this speaker would be making the enclosure, with all those large panels and not-right-angle cuts.

Ralf
 
Good that you've made a decision! They are still BIG speakers though, so I hope the WAF is still there afterwards! (This would be true whichever one you built).

I wouldn't be too concerned over what type of solder you use. Be more concerned that you have a solder sucker, solder wick, croc clips and copper braid for drawing heat away from travelling to components. Crossovers though, are not PCBs, they are large, generally wired point to point. In some cases, instead of solder tag strips as Troel used, I have seen folk use an electrical junction terminal block (the cheap plastic type used in household electrics, particularly light fittings). This would potentially mean no soldering at all.

Of course, a solid conductive connection without parts that may loosen is best and in this regard, it's better to use the solder tags Troel has in the crossover pics. Soldering at these points is good because you can easily put a crocodile clip attached to some copper braid on the solder tag and another on, say, a capacitor lead you're connecting, to act as a heat sink while soldering. You don't need a fancy soldering iron or fancy solder when the connections are this big and easy to work with. You might really only get fancy if you wanted to do it carefully on a PCB, but you would not get any benefit having a crossover on a PCB - if anything that would be worse.

I know PCBs look pretty and nice and tidy, but the crossovers as in the DTQWT pics on Troel's website is the best way to do it.

If this is your first speaker project, I would be more concerned over having the tools for holding such a large cabinet together when you've glued panels together. The cabinet building is the harder part.

You said: "I would like to use the laminated boards on all exterior panels"

I missed this bit in skim reading his build. I hope this is not flooring laminate which I would steer clear of. Mr Lampizateur (www.lampizateur.eu) did a similar thing and reported it was a nightmare to work with - looks simple, but isn't. If the laminate top is plastic, its going to go wrong, if it's all real wood, it will cost a lot (at least here in the UK it would). I assume by laminate, we are actually talking about wood veneer sheets?

There are a few places around where you can get real wood veneers that will be prefinished in a similar style to Ipppe's work. I am *NOT* talking about the cheap plastic stuff on low end speakers to make chipboard supposedly look like Primavera Oak! You're in the USA so I doubt there's any trouble getting hold of decent stuff. A quick google and something like Rosebud Veneer - Lexington, Kentucky, USA may do. Their's are just wood sheets, you can specify thickness required and try to get matched pairs etc. Not expensive either.

Why all wood veneers? Well, you can glue them on, in some cases they have a good glue preapplied and you peel off a backing. You fit it, if you use barely/just oversized sheets, you can trim the edges, sand down if needed and finally treat/wax/stain/seal/clearcoat as required. With a design like Ipppes - if I was reproducing that - I'd get a decent two part epoxy for the black panels (seal wood surface to make it able to receive paint properly) and go for a piano gloss black finish. The front, I would find decent butchers block and do exactly what he did in treating it and waxing it.

Lastly, for where veneered edges meet at vertices and are trimmed/sanded to a perfect edge, I would use a little wood glue - the kind that dries like wood and is sandable. It will make the edges look more like the corner of a solid block rather than two different laminates. It is better to order separate sheets that appear to join at these points. The aforemention veneer supplier looks like they could accomodate that. Possibly build a box cabinet the size of the speaker, order double the required veneers and have a trial run first on the cabinet, so you get an idea of how to work with it and what you can do - play with different sides of the box to see what looks you can achieve. Then, when you have confidence in your workmanship and know exactly how you plan to finish the actual speakers, go for it with the utmost care and passion. It's the love you put in to creating something that will ultimately speak 'quality'.
 
Basically you need to be fast and make good joints. A solder with a small component of silver (2%) is of audio quality.

Silver content solder slows down how fast you can work and has a higher melting point. I would ensure that my hookup cable was in contact with the component lead at the solder tag through-hole, and use normal leaded solder to be fast. Electricity takes the path of least resistance, so if your capacitor lead rests against your hookup lead as they go through a solder tag, a quick dab of normal solder around the outside to fill the tag hole is not going to make any difference.

If you tie/bend a component lead around one half of a brass solder tag, and do the same with your hook up lead on the other side of the through-hole, then solder the hole shut, that might make a minor difference for the worse (making that solder tag junction a point of increased resistance), but I wouldn't get wrapped up over it, it's unlikley to be audible.

I cannot think of anything worse than somebody doing a first build on some excellent speakers and then afterwards getting concerned about what solder they used rather than enjoying the result. They wouldn't be able to tell the difference. There are plenty of excellent, expensive esoteric speakers out there people would love to own that actually have normal leaded solder in.
 
InfiniteGain,

Sorry, for the confusion. I meant Butcher block, as in taking separate narrower boards and side gluing to make a larger surface which is less prone to warping over time. I used the term laminate, because I saw it used in a previous thread here.

I do have experience wood working. I also have a pretty decent woodworking shop for DIY. I have built several subwoofer cabinets and furniture pieces in the past. I have worked with veneering as well. Ippe used side glued pine boards for the front baffle. I would like to carry that to all exterior panels in either cherry or mahogany, for 2 reasons. One is I hate the amount of dust that MDF creates. Every time I work with it, I spend weeks trying to get all the dust out of my shop area. All I have is a shop vac that attaches to my table saw. Not to mention routing it. Second, I believe after cost and time spent is factored in, the cost of using solid or veneer will be pretty close. I need to research this a little more. I have exchanged emails with Troels and he has told me this will work fine.

A local shop, Colonial Hardwood in Springfield VA, actually stocks pre-veneered 3/4 inch MDF specifically for speaker projects. They have had a lot of requests. In the end, if I can't get the cuts right, they will cut it for a small set up fee.

The WAF isn't an issue. The DTQWT is actually only 3 inches wider and 2 inches taller than my current Paradigm speakers.

I have no experience soddering, so thank you for your advice. I will be searching threads here and requesting more help as I go along. I have a couple of house projects, I need to finish before I start. I expect to hopefully get started in a couple of weeks.
 
If you are au fait with woodworking then that's excellent. I don't have all the tools, so have to resort to tricks and outsourcing some to people I know who work with wood all day.

Soldering isn't tricky with this size of stuff. You will be fine. It's good to hear Troel is helping you too.

I am currently getting prices out of Jantzen for the kit or just the drivers. What they sent me looked a little on the high side. The US dollar figure at current exchange rates is £640 (what I would expect) while their Euro price, which is what they will charge me, translates to £840. I find the £200 difference hard to account for. I may try and build it with alternate electronic components of equivalent electrical value.
 
I was a little surprised by the price as well. The price I was quoted for the DTQWT kit with Superior Z-caps and wax coils was $1050. The shipping was an additional $210 U.S.. The Eminence drivers would set me back an additional $109 each, so $436 total with free shipping. So ~$1700 for components before cabinets are factored in. I may try to see if they have other shipping options, I am not in a hurry.

I actually spent a night, pricing out the individual components from different companies on parts express and madisound. If I were to use clarity SA caps or Jantzen purchased locally, the cost would be higher. The Audax tweeter I found for $75 from madisound. The only source for the midrange is parts express and they quoted a price of around ~$375 each if I remember correctly. So the kit is a better deal.

Looking at the pdf that Jantzen Audio sent, it seems like a pretty complete kit. This is my first speaker project, do you see any items missing from the kit that I will need to buy other than damping material?

I do feel better using the exact parts that Troels recommends. Plus for anyone considering this project, I have emailed Troels repeatedly and he always answers very quickly. Make sure your subject line is very clear for his spam filter.
 
Hi,

I think you are right that the price differential is partly due to tax. UK Value Added Tax is 17.5%. If you take a taxless price of £640, then VAT will not take you anywhere near the £840. Tax would need to be over 30%.

I contacted the UK Jantzen distributor about the drivers alone. The JA8008 is £411.30 a pair, and the TW034 with waveguides is £263.20 a pair. Together, no shipping, that's £674.50. That's serious outlay. I'll be pricing up the other components from alternative makers later tonight.

As for the midrange, this is a SEAS driver built to Troel's spec. It's based around CA21REX and I cannot find a price on that anywhere. The Audax driver is 80 Euros or £72, so £144 for both versus £263.20. Looks like I may buy the JA8008 from Jantzen, get the Audax myself (and do any required tweaks) along with the rest of the crossover components. If I could find the SEAS CA21REX and mod it myself, I'd be happy with that.
 
It does look like the Jantzen kit is my cheapest offering. I am not living in my main home presently so cannot cut the wood at home. I found a cutting house that will charge £210 for all the panels, chamfers/bevels and routing. Rather good I think.

I have found the Deltalite 2510's will cost me a shade under £300 delivered inc taxes.

I will order a new HVLP as I would like to do a piano gloss. This and associated materials will bring total price to around £1750 which is acceptable for what I believe will be the sort of performance normally associated with speakers in the £10k region. I formerly had Wilson Benesch ACT Ones and still currently own Wilson Benesch Orator's which have a good response down to 30Hz. I expect the DTQWT to walk over them. The Orator's only use a Scanspeak 8545 midwoofer, a Scanspeak 9300 tweeter and have bass tuned through a vented plinth. This transpires to be round port in an otherwise sealed cabinet with about 7 or 8 mm to the plinth. The techno blurb calls the low frequency tuning as "bessel alignment of fourth order reflex, double chamber differential tuning".

I'm not sure if WB's blurb is correct. I'm no expert, but the following shows only 2 designs having a port to outside the cabinet.

DCR

Although WB mention "fourth order reflex" I don't think they mean 4th order passband. Looks more to me that they have implemented a DBR Fostex style - but does anybody know for sure?
 
Hi,

I live in Denmark (as Troels) and know Troels. Heres some comments from me.

Troels has build many speakers and have a lot of experience. There are still some things he need to look into, but he is better then most DIY.

I dont like the custom Seas driver used, as the suspension and the moving parts doesnt match (they resonate as seen on the impedance of the driver). We have to go many years back to find drivers with this error.

He also finds the delay between the drivers by using the stepresponse, wich is impossible. You have to calculate it from the Groupdelay to get anything accurate. Did you know he earns money for each sold kit?

That said, hes a good designer.

Zaphs graphs is useless. Ranging from 35 to 125dB ?!?! you have NO resolution. Everything would look good... 2dB ripples still looks like a strait line.
I cant find anything about how he finds the Delay needed on the larger drivers. Does anybody know?

His enclosure calculations are based more on frequencyresponse, than considering the time domain. His 18W SS constructions are BOOM boxes...

Zaph made a great comparable test of drivers. Thats all i would use from that website, sorry.


The above is not to disrespect any of them. Just my honest opinion.


I would recommend a 3way SB construction from Kasper (Kappen) (also living here in Denmark). His a very talented engineer with close contact with Ulrik S. from SBacoustics (earlier Scanspeak engineer).

See his construktion build here :
HIFI4ALL.DK Forum: Toebs 3vejs SB DIY

You can buy it here, by contacting Kasper from his webshop, the kit is not on the webpage yet, but will be soon :
Rabu Acoustics

An externally hosted image should be here but it was not working when we last tested it.
 
InfiniteGain,

I've reached the same conclusion. I am delaying the start of my speaker project until my shop gets cleaned up. No space right now. I am veneering a subwoofer cabinet today. It is taking up my work bench right now. The weekend after Easter will probably be when I start. Please let me know how your project goes. I am learning as I go.
 
Hi,

I live in Denmark (as Troels) and know Troels. Heres some comments from me.

Troels has build many speakers and have a lot of experience. There are still some things he need to look into, but he is better then most DIY.

I dont like the custom Seas driver used, as the suspension and the moving parts doesnt match (they resonate as seen on the impedance of the driver). We have to go many years back to find drivers with this error.

He also finds the delay between the drivers by using the stepresponse, wich is impossible. You have to calculate it from the Groupdelay to get anything accurate. Did you know he earns money for each sold kit?

That said, hes a good designer.

Zaphs graphs is useless. Ranging from 35 to 125dB ?!?! you have NO resolution. Everything would look good... 2dB ripples still looks like a strait line.
I cant find anything about how he finds the Delay needed on the larger drivers. Does anybody know?

His enclosure calculations are based more on frequencyresponse, than considering the time domain. His 18W SS constructions are BOOM boxes...

Zaph made a great comparable test of drivers. Thats all i would use from that website, sorry.


The above is not to disrespect any of them. Just my honest opinion.


I would recommend a 3way SB construction from Kasper (Kappen) (also living here in Denmark). His a very talented engineer with close contact with Ulrik S. from SBacoustics (earlier Scanspeak engineer).

See his construktion build here :
HIFI4ALL.DK Forum: Toebs 3vejs SB DIY

You can buy it here, by contacting Kasper from his webshop, the kit is not on the webpage yet, but will be soon :
Rabu Acoustics

An externally hosted image should be here but it was not working when we last tested it.


Syncroniq,

Thanks for your response. I am new to speaker building, so I hesitate to argue some of the more technical aspects I see debated on these forums. In particular, I do not fully understand the different measurement techniques I often see given here or the arguments why one is better than the other.

I have searched for several months for a 3 way floor standing speaker. These speakers will be used in a dedicated home theater room, as well as critical listening, so low end response is important to me.

In my search I found some designers who were referenced over and over again. These being two of the most prominent. I have not found anything to say their designs are poorly made. I have seen some write ups that fault a specific characteristic or use of a particular driver. My understanding is that those were personal preferences and not a bad design.

My concern is with your comment about the midrange driver that Troels had SEAS build. My understanding is that SEAS is very well known for the quality control of its drivers. I have never seen them in person. You say that the moving parts resonate with the suspension which causes the impedance plot to spike. I understood the impedance never drops below 6 ohms. When you say resonance, are you speaking of an audible distortion? I understood the light weight cone is able to be extremely accurate at high spl levels. Are you saying it isn't?

As for a designer receiving a portion of the sales of a kit, that seems reasonable to me. I am a professional with 3 children and I barely have time to surf the forums occasionally. The only time I get to build is if I decide to stay up after the kids at night. So a portion of sales, to cover design, experimentation, and website content seems only fair.

If anyone else, who is familiar with these designers, would care to respond please do so. I haven't purchased a kit yet.
 
usmcjlp,

I think Syncroniq's comment on we can see the driver resonate at the impedance spike is a bit misleading. He needs to explain this a bit more for you to understand what he means. To illustrate, I'll borrow some explanation from Eminence's explanation of speaker parameters (Thiele/Small parameters):

Fs This parameter is the free-air resonant frequency of a speaker. Simply stated, it is the point at which the weight of the moving parts of the speaker becomes balanced with the force of the speaker suspension when in motion. If you've ever seen a piece of string start humming uncontrollably in the wind, you have seen the effect of reaching a resonant frequency. It is important to know this information so that you can prevent your enclosure from 'ringing'. With a loudspeaker, the mass of the moving parts, and the stiffness of the suspension (surround and spider) are the key elements that affect the resonant frequency. As a general rule of thumb, a lower Fs indicates a woofer that would be better for low-frequency reproduction than a woofer with a higher Fs. This is not always the case though, because other parameters affect the ultimate performance as well.
- Eminence website - supplied for information purposes only, no copyright infringement intended and data is already public domain.

At the Fs (free resonance) of a speaker driver, the impedance spikes sharply. This happens with all drivers, not just the Seas Troels modified. Plus, I would err on caution as Troels modifications are to compensate for what he perceived as the SEAS failings and Syncroniq does acknowledge him as a good designer. Are Syncroniq's comments on the base SEAS or the derived JA8008? I'm not sure.

Perhaps Syncroniq means that at the SEAS resonance frequency, it's resonance causes it to exceed its Xmax (maximum linear excursion) - a point where the driver is pushed enough that it leaves the area where the magnet can maintain linear control over the driver (thus introducing distortion and beyond proper amplifier control for a few milliseconds). However, I don't know if that is true or where he got that information. Normally you would measure this yourself as it isn't necessarily inferred from driver parameters.

I have heard that this driver (original SEAS) is a little bass light, but in this instance we are supporting that bass using the tuned horn enclosure and two big 10" high efficiency drivers on the back of the cabinet and with measurements confirming a good in room response down to 24Hz. The recorded in room response is what tells us the truth of the matter. Agreed, it does not inform us of distortions, but it appears to me Troels is using MLSSA software (amongst other software) and so can measure distortions if required. Perhaps, you could ask him?
 
Last edited:
Hi,

I live in Denmark (as Troels) and know Troels. Heres some comments from me.

Troels has build many speakers and have a lot of experience. There are still some things he need to look into, but he is better then most DIY.
...

Did you know he earns money for each sold kit?


...

The above is not to disrespect any of them. Just my honest opinion.


I would recommend a 3way SB construction from Kasper (Kappen) (also living here in Denmark). His a very talented engineer with close contact with Ulrik S. from SBacoustics (earlier Scanspeak engineer).

See his construktion build here :
HIFI4ALL.DK Forum: Toebs 3vejs SB DIY

You can buy it here, by contacting Kasper from his webshop, the kit is not on the webpage yet, but will be soon :
Rabu Acoustics

An externally hosted image should be here but it was not working when we last tested it.

Forgive me if I'm wrong, but I can not help myself from having the impression of your post being a kind of unfair advertisement.
I believe that Troels started as purely enthusiastic DIY-er and after a huge number of successful designs he began making some money from his work in this field. That said, is your friend Kasper giving it for free?
 
Hi,

All units have a resonant frequency, so thats not what im talking about.

Take alook at this picture with the impedance plot of the driver/system.

An externally hosted image should be here but it was not working when we last tested it.


All the small ripples between 120Hz and 500Hz are bad resonance between the suspension and moving mass. You can even see it on the phaseplot too. That tells that the drivers parts are not matched well. I belive Troels designed the driver himself, so you cant blaim it on Seas. They only produce it. They make some of the best drivers.

No my friend does not give away the design, he has a shop as stated on his website. Troels on the other hand does not say anything about him making money on the projects. So when he says "the best project ive made sofar", well... your level of critic would be wrong.

But, i belive Troels is a good speakerdesigner. I think he should let Seas design his units though. And i think he should let people know he makes money on the projects sold.

Of the two projects, i would build Zaphs design in a smaller box. Becourse i belive the drivers are alot better in that design.

I dont think you can buy my friends project, duo to distributor limitation (he cant sell to countries were theres another SB distributor. But you could ask for the design and hope he will give it away. Its a better project then both of the above.
 
Last edited:
Hey,

Im not good at english, i know. Im sorry for that.

The peak at 900Hz is the loss of the suspension on the active coneareal. Above that frequency, the units simply doesnt use the susprension any more. I dont know how to explain it, but all units have it, so thats ok.

The other ripples are small resonances. What about resonances is it you dont understand? Again im sorry that my english is bad. I wish i could be more precise. English is not my first language.

I have a close contact with many engineers, both vifa, peerless, scanspeak and sb. I have designed motorsystems, BL linear coils systems, made pro.speakerbuilder.dk, http://graph.flexunits.com:8080 for audiotechnology, many kits for dealers and companies, and worked as a distributor for Tymphany some years back. I have my own Clio at home with LspCAD and have had for about 5 years now.

a small portion of my speakers are displayed for free at Speakerbuilder but most of my work is for privates with very special designes. I dont see any reason for displaying them, as they are impossible to copy. I think i have made the first 25-30 speakers in 2009 in my spare time.

Right now, im working on a SBAcoustic 2½ way design :
http://forum.speakerbuilder.dk/forum_posts.asp?TID=231

That would be my answer. Please let me know if i answered in east, and you asked in west.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.