ATC - replace or DIY project?

[Jas0_0]

Ok... so in February I bought some ATC active monitors. Yes, yes, I saw Amir’s review of the SCM19s. You can stop sniggering.

Mine are hybrids with the cabinet and drivers of the SCM20ASL mk 1 and the amp packs of the T16. They were club DJ monitor speakers, so they’re well used, and I got them for a very very good price (so the damning ASR review didn’t hurt too much). They also added enough bass extension, scale and detail over my previous Proac/Croft (yes, yes, valves, I know, I wish I’d discovered ASR sooner...) combination for me to sell both.

I had intended to send them to ATC to have the bass drivers re-coned and the amp packs upgraded to full 20ASL spec. But then I read the ASR review and wondered if it was worth it. My first reaction was that I should just cut my losses, sell the ATCs for what I could get for them and buy some Genelecs instead.

But then I wondered if I could use the ATCs as the basis of a great DIY project... with a digital crossover, class D amps and new bass drivers - possibly the Purifi ones. These ATCs have a sealed, curved, cast aluminium enclosure and come with decent Seas tweeters - both plus points. So if I had new front baffles made to take the Purifi drivers, used two Hypex NC400 amps a side fed by my RME ADI-2 Pro FS (I can buy some good products...) in multichannel mode, and then used software on my Mac to perform crossover and DSP, could I create a good speaker?

Or am I being arrogant to think that my Frankenstein speakers could come even close to established designs like Genelec?

 

[AudioStudies]

I think you would need to verify that the enclosure volume would be a good match for the driver characteristics.

 

[maverickronin]

Or am I being arrogant to think that my Frankenstein speakers could come even close to established designs like Genelec?

Matching their 2-ways should be plenty doable if you're willing to put in the work.

There are no off the shelf parts to replicate The Ones though.

 

[Jas0_0]

I think you would need to verify that the enclosure volume would be a good match for the driver characteristics.

Thanks, I am completely new to this - is there a rule of thumb, or maybe a set of equations that I should use to determine whether a driver is appropriate? I believe the ATC enclosure is 20 litres. The only reference I could see to volume in the Purifi driver specs was “Eq. volume, Vas - 26L”. Is this relevant?

 

[Jas0_0]

Matching their 2-ways should be plenty doable if you're willing to put in the work.

Thanks, this is encouraging. Though I wonder if I'll ever be able to achieve the off-axis performance of the Genelecs, with their wave-guided tweeter.

Assuming I can, I am really new to this and I wonder if you could help with a few more questions:

Firstly what measurement equipment will I need? I have a UMIK mic and REW, and could use the garden and a stepladder to perform quasi-anechoic measurements. Would this be sufficient?

Secondly, would the Hypex NC400 modules and power supply be happy mounted inside the ATC enclosure, perhaps coupled in some way to the heat sink?

Thirdly, aside from high/low pass, phase correction, EQ and room correction, are there any other filters required/advised for a digital active crossover?

Lastly, are there any other 6.5" woofers, aside from the Purifi, that are highly regarded for DIY projects?

Is there anything else I should know before embarking on this?!

Thanks!

J

 

[AudioStudies]

Thanks, I am completely new to this - is there a rule of thumb, or maybe a set of equations that I should use to determine whether a driver is appropriate?

There are equations, and it is a bit more complicated than I can fully explain here. I recommend obtaining a book on speaker design that explains the Thiele-Small parameters (TS parameters), and how they relate back to the enclosure that a driver is installed within. It may be possible to find this info online also. For a given driver and its associated TS parameters, the frequency response will vary with enclosure size (internal volume of the enclosure) and also whether any absorbent material is stuffed within the enclosure. The Vas is relevant but not until you understand how it relates back to the aforementioned things.

 

[Chrise36]

Ok... so in February I bought some ATC active monitors. Yes, yes, I saw Amir’s review of the SCM19s. You can stop sniggering.

Mine are hybrids with the cabinet and drivers of the SCM20ASL mk 1 and the amp packs of the T16. They were club DJ monitor speakers, so they’re well used, and I got them for a very very good price (so the damning ASR review didn’t hurt too much). They also added enough bass extension, scale and detail over my previous Proac/Croft (yes, yes, valves, I know, I wish I’d discovered ASR sooner...) combination for me to sell both.

I had intended to send them to ATC to have the bass drivers re-coned and the amp packs upgraded to full 20ASL spec. But then I read the ASR review and wondered if it was worth it. My first reaction was that I should just cut my losses, sell the ATCs for what I could get for them and buy some Genelecs instead.

But then I wondered if I could use the ATCs as the basis of a great DIY project... with a digital crossover, class D amps and new bass drivers - possibly the Purifi ones. These ATCs have a sealed, curved, cast aluminium enclosure and come with decent Seas tweeters - both plus points. So if I had new front baffles made to take the Purifi drivers, used two Hypex NC400 amps a side fed by my RME ADI-2 Pro FS (I can buy some good products...) in multichannel mode, and then used software on my Mac to perform crossover and DSP, could I create a good speaker?

Or am I being arrogant to think that my Frankenstein speakers could come even close to established designs like Genelec?

First is what you dont like about their sound?You dont have to change anything and the bass driver is a low distortion unit just compare it with any other speaker measured here.Making decisions just from a measurement and that was another model not measured in room is not correct imo.These are high quality speakers their cabins are better almost twice as heavy as other bookshelfs without resonance like the Revels made in UK not Asia etc.My passive scm20 measure totally different in my room. I would use them with a sub from 80hz and up and a bit of eq according to your taste.If you want to experiment you can start with the hypex plate amps with integrated dsp.From there i would go to three ways.

 

[maverickronin]

Though I wonder if I'll ever be able to achieve the off-axis performance of the Genelecs, with their wave-guided tweeter.

Without a waveguide you'd need to cross over the tweeter really low, before the woofer starts beaming. IDK what tweeter that has, or if it's capable of that so you might need new tweeters too.

Firstly what measurement equipment will I need? I have a UMIK mic and REW, and could use the garden and a stepladder to perform quasi-anechoic measurements. Would this be sufficient?

That plus splicing in ground plane measurements for the bass.

Secondly, would the Hypex NC400 modules and power supply be happy mounted inside the ATC enclosure, perhaps coupled in some way to the heat sink?

I'm sure that's possible with enough custom mounting parts, but it will probably eat more volume than the stock electronics. Keeping it external would be easiest. I've seen people make boxes for the Hypex Fusion plate amps and keep them external. That's probably a good choice.

Thirdly, aside from high/low pass, phase correction, EQ and room correction, are there any other filters required/advised for a digital active crossover?

Off the top of my head there's also baffle step compensation and level matching between the woofer and the tweeter.

Is there anything else I should know before embarking on this?!

It's going to be a ton of work if you've never done it before. You're going to need to do a lot of reading, learn how to use some simulation programs to keep yourself from wasting money on unsuitable parts, keep dragging your speaker back and forth to make changes and remeasure, etc, etc.

 

[Jas0_0]

Thanks @maverickronin - really helpful info. Hypex plate amps seem like good and more cost effective alternative to nc400s. Seems I have a lot to learn and a lot of work to do. The big question I guess is will it all be worth it? On the plus side, I'll have totally unique speakers that I can proudly look at and think "I did that". But after all this time and investment, will I really end up with a speaker that sounds considerably better than an off-the-shelf Genelec 8340, or similar?

 

[maverickronin]

The big question I guess is will it all be worth it?

If you value the learning you'll need to do along the way, then yes. This is something to do because you think it's fun, not just trying to save money. Counting the time you spend on it as "work" will make it considerable more expensive.

But after all this time and investment, will I really end up with a speaker that sounds considerably better than an off-the-shelf Genelec 8340, or similar?

It's probably not going to be better than one of two way Genelec's but with enough work it should be possible to make something on par with them.

 

[Jas0_0]

I’ve found a couple of books on speaker design on Amazon. Is there one that’s considered the speaker builder’s bible, that covers active monitors?

 

[maverickronin]

I’ve found a couple of books on speaker design on Amazon. Is there one that’s considered the speaker builder’s bible, that covers active monitors?

I'm not sure if there are any ones specifically on active monitors.

 

[AudioStudies]

I’ve found a couple of books on speaker design on Amazon. Is there one that’s considered the speaker builder’s bible, that covers active monitors?

I am not sure about any that cover active monitors. It has been quite some time since I have read them, that I can't recall which are the better ones.

 

[ctrl]

Without a waveguide you'd need to cross over the tweeter really low, before the woofer starts beaming. IDK what tweeter that has, or if it's capable of that so you might need new tweeters too.

Without knowledge of the baffle dimensions it is hardly predictable.

With a baffle width of 0.22-0.26m a dome tweeter without waveguide and a 6''-7'' woofer with a rather high crossover frequency (>2kHz) will probably give a more even power response.
At a low crossover frequency the dome tweeter will probably radiate very wide between 2-3.5kHz due to edge diffraction and lead to a peak in power response.

It's going to be a ton of work if you've never done it before. You're going to need to do a lot of reading, learn how to use some simulation programs to keep yourself from wasting money on unsuitable parts, keep dragging your speaker back and forth to make changes and remeasure, etc, etc.

I can confirm that. If it is really your goal to compete with professional loudspeaker developments, then you should expect months of studying theory and getting used to simulation software.
For example: http://www.randteam.de/

But after all this time and investment, will I really end up with a speaker that sounds considerably better than an off-the-shelf Genelec 8340, or similar?

I don't think you're gonna get that. The first DIY loudspeaker development is unlikely to reach such a high standard.

If it were that simple, there would be no need for professional loudspeaker designers at all, as anyone could become an expert in a few weeks - which is clearly not the case.

 

[AudioStudies]

I don't think you're gonna get that. The first DIY loudspeaker development is unlikely to reach such a high stand

I agree with CTRL. Nevertheless, it may be a fun endeavor and great learning experience. The first step is to study enough theory about TS Parameters and Volume Enclosures to determine what the frequency response will be for your intended driver and enclosure. You may determine that the driver and enclosure are not a good match. Or conversely, they may be a great match, or something in between. Even if you decide not to build the speaker, the knowledge of speaker design will be of benefit when evaluating loudspeakers for purchase, or subsequent DIY designs. I wouldn't worry too much about finding a design book specifically for active monitors. There are a lot more important things on your plate before that. Public library may have speaker design books also to supplement info from those that you purchase. Its been a while, but I really enjoyed learning about the theory.

 

[Jas0_0]

Thanks everyone for your input. I think this will be worth trying. Already having fun looking for midbass drivers that will work in a sealed enclosure.

I think I will attempt this project in stages, first replacing the AB amps with class D and going from analogue to digital crossovers. Then once I’m comfortable I’ve got the hang of that, I’ll read up about replacing the front baffle and drivers.

Yesterday I dismantled one speaker and discovered the amp pack is huge - much bigger than two nc400s and a SMPS1200 combined, so I think I’ll go the internal route rather than using plate amps or building outboard boxes.

Also discovered that the enclosure damping rubber has come unstuck and sealing gaskets are in poor shape, so that will be another short term fix.

Final thing will be cosmetic overhaul. Think having the aluminium enclosure shot blasted and clear coated, and lacquered, CNC-cut ply for the front baffle will give them a chic industrial look!

 

[b1daly]

How do they sound to you, are you unhappy? Frankly many flaws in speakers can be helped with hi quality dsp parametric EQ. depending on your setup this can be a real pain to implement, but much less than the options you are conidering!