DIY Harbeth 40.1

Think so? Not sure to be honest, haven't got that far :D ? I'm burning up the youtube wires. The har-beth's are 6 ohm, I want mine to be 8 ohm, so i'll have to adjust things accordingly.

I'm having fun though.

If I get any closer, I keep getting this strange feeling I'm gonna get a visit from Mi6 soon :D
 
Last edited:
Do yourself a favor and order a copy of the Loudspeaker Design Cookbook off of amazon. It will help you a lot with the design of crossovers, enclosures and driver choices. It's a great book.
 
It's a fairly "thick" read. I had to do it in parts multiple times, but I did learn something!

Good luck with the project. Sounds like you're going for it come hell or high water. I respect the enthusiasm. If you enjoy the process the end result is only a fraction of the overall enjoyment!
 
I have built some clones of the Fried T-line and aperiodic designs over the years.
These designs use a series resonance crossover circuit.

Getting the crossover and enclosure damping material correct on a 2 way can take about a month of critical listening and taking measurements.
It is a good idea to have plenty of alligator clips handy for comparing different cap and resistors values to obtain the best response.
 
I have built some clones of the Fried T-line and aperiodic designs over the years.
These designs use a series resonance crossover circuit.

Getting the crossover and enclosure damping material correct on a 2 way can take about a month of critical listening and taking measurements.
It is a good idea to have plenty of alligator clips handy for comparing different cap and resistors values to obtain the best response.

Probably be handy to have a pair of 40.1 to compare to, as well.
 
Some of the Stereophile reviews were pretty detailed, concerning measurements, which should at least help figuring out crossover frequencies. Maybe that'll help?

http://www.stereophile.com/standloudspeakers/1008har/

Also, I found this Chinese web-site not too long ago. They carry Vienna Acoustics kits, and drivers. I wonder if they might be a worthy Harbeth substitute? Vienna Acoustics aren't too shabby either!

http://www.aliexpress.com/item/2pcs-pair-Vienne-7-spider-cone-midbass-woofer-seas-vifa-scanspeaker-morel-eton-hivi/1355702528.html

http://www.aliexpress.com/af/vienna.html?ltype=wholesale&exception=&maxPrice=&isFreeShip=n&SearchText=vienna&shipFromCountry=&isUnitPrice=&isFavorite=n&isAffiliate=y&needQuery=y&shipCompanies=&isOnlyAffiliate=y&d=y&origin=n&shipCountry=us&isAtmOnline=n&maxQuantity=&isBigSale=n&CatId=518&isViewCP=y&pvId=&isOnSale=n&isRtl=n&minQuantity=&minPrice=

Ya' gott'a start somewhere! :thmbsp:

Good Luck!
 
Last edited:
I've built many dozens of DIY speakers. Perhaps, even a dozen dozen. :)

I don't think you are going to recreate the sound exactly, but there are things you could try to determine in order to make a DIY speaker have a better chance at success. Some time spent determining the attributes you are looking for where the 40.1s excelled would be a good start. If you can do that, there are people here who may be able to help correlate what you are describing to an objective performance attribute/measurement.

For me, while I find the 40.1's to be fine speakers, they are too inefficient to fully capture my attention as an enthusiast or builder. I'm sure I'm suffering from expectation bias based on the very low sensitivity, but when I heard them, they didn't sound like "live" music, though they did sound quite nice. I think I'm overly sensitive to power compression, however. If I didn't occasionally like to turn the volume up, I may be very happy with them.
 
Last edited:
..., but when I heard them, they didn't sound like "live" music, though they did sound quite nice. I think I'm overly sensitive to power compression, however. If I didn't occasionally like to turn the volume up, I may be very happy with them.

This is, I'd opine, a common feature of many loudspeakers nowadays -- in some cases, it can be overcome with the use of prodigious amounts of amplifier power -- but, by and large, some of those "nice" sounding loudspeakers just don't sound real (i.e., I am empathetic).
 
Well I'm going to go against the grain here. If the OP wants to take on this challenge more power to him. I think he understands the risks involved and apparently he feels that the rewards are greater than those risks.

So let's offer him some encouragement and our well wishes rather than tell him what he can't do. That attitude is counterproductive and benefits no one. Good luck man! Hopefully, someone with DIY expertise will step in and offer some helpful suggestions.

cubdog

+1 If you have the desire,go for it
 
I dunno. I don't think he does understand the risks or what's involved.
I think he should attempt something less stellar to hone his skills and once he's successfully accomplished that increase his scale.

One can say they want to win the Tour De France but don't own a bicycle and have never ridden one, but I think encouraging one to do so is folly on both sides. Stuff like this happens on TV, but not too often in real life....
 
I've built many dozens of DIY speakers. Perhaps, even a dozen dozen. :)

I don't think you are going to recreate the sound exactly, but there are things you could try to determine in order to make a DIY speaker have a better chance at success. Some time spent determining the attributes you are looking for where the 40.1s excelled would be a good start. If you can do that, there are people here who may be able to help correlate what you are describing to an objective performance attribute/measurement.

For me, while I find the 40.1's to be fine speakers, they are too inefficient to fully capture my attention as an enthusiast or builder. I'm sure I'm suffering from expectation bias based on the very low sensitivity, but when I heard them, they didn't sound like "live" music, though they did sound quite nice. I think I'm overly sensitive to power compression, however. If I didn't occasionally like to turn the volume up, I may be very happy with them.

This is, I'd opine, a common feature of many loudspeakers nowadays -- in some cases, it can be overcome with the use of prodigious amounts of amplifier power -- but, by and large, some of those "nice" sounding loudspeakers just don't sound real (i.e., I am empathetic).

A friend of mine demo'd the 40.1 for his mates. They all preferred the tube-era JBLs and Lowthers over the Harbeth. His take on it was that hobbyists tend to prefer more idiosyncratic sound over flat & neutral sound. He sold the Harbeth soon.

But I know he doesn't have very powerful amps, being a high-end vintage tube guy. The most powerful would be TOTL Leben separates in his collection. It could have been the insufficient amplification. Having said that, isn't the Harbeth dude one of the amps-is-amps guys?
 
As a few people have said, part of the "magic" of the Harbeth design is that the cabinet is thin walled and adds to the sound output of the speaker. Most designs use an inert cabinet (thick walls, internal bracing, etc) to take the cabinet out of the equation and deal only with the output of the drivers. Harbeth uses a thin wall that has its own resonances and additions to the overall sound. Duplicating that without actually copying the actual plans precisely will take a lot (and I mean a LOT) of measurements. There is the chance that you can get lucky but there is a reason why the overwhelming majority of companies and DIY builders work to keep the cabinet as inert as possible.

Step 1: the preliminary sketch...

Read up on edge diffraction/re-radiation a little. Moving the mid and tweeter off the vertical centerline will help to smooth out the response of the speaker overall.
 
I've noticed all the Har-beth's are designed at 6 ohms. I wonder why? I've also noticed that many many British designed speakers are not very efficient, below 90db. They build and use very stiff cones. kevlar, etc. I imagine if you have a very clean 50w/ch amp, i believe it will suffice driving any of the Har-beth's. Here's an example of the HL5's with a a Red Wine Audio 57 amp (57 watts RMS per channel into 8-ohms). It sounds pretty good to me! https://www.youtube.com/watch?v=xu9w7S8Rzy4 and from the video it's almost at half volume...I myself like neutral (in your face) detailed speakers.

Studying the 40.1's again, they are a big cabinet speaker, I've sourced a very comparable KEF subwoofer for the bass driver, but I've been thinking of tackling the hl5's. Not that the task would be any easier, merely for scale :D

I've designed and built furniture before, that is not the challenge, the problem is the crossover, who can help with that? Anyone know any links to some simple solutions or calculator's that wouldn't require a physic's degree?

The loudspeaker cookbook is one resource, any others?
 
Last edited:
I've noticed all the Har-beth's are designed at 6 ohms. I wonder why? I've also noticed that many many British designed speakers are not very efficient, below 90db. They build and use very stiff cones. kevlar, etc. I imagine if you have a very clean 50w/ch amp, i believe it will suffice driving any of the Har-beth's. Here's an example of the HL5's with a a Red Wine Audio 57 amp (57 watts RMS per channel into 8-ohms). It sounds pretty good to me! https://www.youtube.com/watch?v=xu9w7S8Rzy4 and from the video it's almost at half volume...I myself like neutral (in your face) detailed speakers.

Studying the 40.1's again, they are a big cabinet speaker, I've sourced a very comparable KEF subwoofer for the bass driver, but I've been thinking of tackling the hl5's. Not that the task would be any easier, merely for scale :D

Neither one would be easy to recreate, probably more likely that it would be damn near impossible to make the same speaker. That's not to say that it would be worse or better, just not the same.
I've designed and built furniture before, that is not the challenge,

I've built loads of speakers and some furniture, I can't stress enough (and in that, reinforce Ray's post) that it will be a significant challenge. The actual construction/assembly (gluing and screwing a box together) is rather mundane but in the case of the Harbeth, the cabinet becomes part of the sound/tuning and works in conjunction with the crossover and drivers. It's going to be no easy task to get things tuned properly. If you are simply looking for the harbeth look, there are few cabinet designs that would be easier to mimic. They are the ultimate in boring boxes, aesthetically, and could be copied by any amateur woodworker. I don't believe that is what you are looking for, however.
the problem is the crossover, who can help with that?

Yes, that will also be a problem. Since you won't find the harbeth drivers, you are designing a speaker from scratch. RayW, dumptruck, GordonW and a handful of others can assist with crossover advice, but it's all going to be done in very general terms as they likely won't have the same parts/cabinets on hand to test.
Anyone know any links to some simple solutions or calculator's that wouldn't require a physic's degree?

At a minimum, you'll need measurement equipment and some software to simulate crossovers. A calibrated mic and REW will get you going. There is a good tutorial with REW, I believe that can help you figure out how to measure the drivers.
The loudspeaker cookbook is one resource, any others?

That's a good one but that elusive physics degree you mentioned earlier would likely help you with some parts of it. It's a great resource, but for someone new, it might be a bit painful to get through.
 
... If you are simply looking for the harbeth look, there are few cabinet designs that would be easier to mimic. They are the ultimate in boring boxes, aesthetically, and could be copied by any amateur woodworker. I don't believe that is what you are looking for, however.
If I may be so bold... I think Mr. N. hit the nail on the head, here. The problem, so to speak, with the Harbeths is that they look so darned ordinary. Not dissimilar to generic three-ways of a generation (or more!) ago. How hard could it be to recreate a pair?

Well... thereby, as I think numerous posts to this thread have illustrated clearly, hangs a tale.
 
I know it won't be an exact duplicate! That can only be possible if I were in the UK on the assembly line and merely assembled one.

I don't think it's soooo special, that it's impossible. Your all reading into it way to much...it's like saying know one can ever figure out how to replicate the secret sauce in a Mcdonalds Big Mac.

My goal here is to produce a DIY version based on the Har-beth design, sound and build. Will it be exact, exact to the precise enth degree? no! BUT, with some careful study, backward detective dissecting, it can come close to that. I want to reproduce that forward neutral sound that makes those Har-beth's so exciting at a fraction of the $13,000 price tag, and who knows, with a little luck, give birth to something new and exciting and maybe frickin better!!!

That's what im hoping :D call me crazy
 
Last edited:
Back
Top Bottom