Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Been gone for a while, is m3 still good for DIY?
New Posts  All Forums:Forum Nav:

Been gone for a while, is m3 still good for DIY?

post #1 of 24
3/1/10 at 4:31am
Thread Starter 
I've been gone from the forum for a few years and recently I took my dt880's out and thought "I really need a decent amp, I just don't have a ton to spend on one."
So I'm going the DIY route and I recall the m3 being a really good amp three or four years ago. Is it still a good DIY amp? or has something better come out since?

I'd also like some suggestions on enclosures. Something that can house both the power supply and amp in one box.
Reply
post #2 of 24
3/1/10 at 5:07am
Yep, it still is and will always be a great desktop amp.
Reply
post #3 of 24
3/1/10 at 5:12am
The M3 is still highly regarded. That said there have been several very good designs in the past several years. I suggest spending some time at the following sites Cavalli, AMB and DIY-Forums/Beezar

Most of the designs have enclosures suggested in the BOM, but i will mention the Nabu it is very affordable and has plenty of room for almost any diy project.
Reply
post #4 of 24
3/1/10 at 5:18am
I'm a great fan of Par-Metal enclosures, though Hammond is the "simple and standard" option as it's super easy to work with and available in most electronic distributors. One thing about that Nabu (which I've used for my personal Beta22 build) - it's steel, RF shielded and freaking heavy!
Reply
post #5 of 24
3/1/10 at 12:20pm
Quote:
Originally Posted by pwrusr View Post
Is it still a good DIY amp? or has something better come out since?
You could do a lot worse than the M3. I can't say whether something better has or has not come out, but it does have a lot more DIY competition, which is a great thing - many options with which you can't go wrong.
Reply
post #6 of 24
3/1/10 at 3:38pm
Quote:
Originally Posted by FallenAngel View Post
I'm a great fan of Par-Metal enclosures, though Hammond is the "simple and standard" option as it's super easy to work with and available in most electronic distributors. One thing about that Nabu (which I've used for my personal Beta22 build) - it's steel, RF shielded and freaking heavy!
Ok, with 7000+ posts is hard to find the pics. I gotta see your b22 in the Nabu. Can you point me to the pics if they are posted? Would love to see external and internals...
Reply
post #7 of 24
3/1/10 at 3:44pm
what Z? - for 32 Ohm m3 would be fine

with 600 Ohms you may not have enough Vswing for dynamic music peaks
Reply
post #8 of 24
3/1/10 at 4:08pm
Sorry for a bit of thread-hijacking OP.

Does the β22 warrant a significant improvement over (well, going discrete is a tempting prospect) the M^3 considering that parts-wise, the former costs at least twice of the latter?
Reply
post #9 of 24
3/1/10 at 8:49pm
with the output biased Class A, having adequate I, Vswing for the can's Z there is some question of whether you might be able to hear differences - especially with controlled testing using level matching and blinding protocols

the M3 will be slightly noisier on the input than the discrete fets of the beta22, but attenuator Z, source resolution or recording noise floor could all be higher than either amp's input noise

the AD8610 is really quite recent, built on a very advanced process by a competent design team - its not likely to have major flaws or be comparable to 70's op amps that could have started the whole "discrete is always better" thing going

keeping in the vein of OPA627 substitutes there is the very new OPA827 from TI which would allow up to +/-18 V supplies for greater headroom with higher Z cans

I don't know if the M3 board layout and gnding could be improved, but the basic topology should be capable of excellent performance with reasonable load Z, sensitivity cans (I would change the cascode ccs to bootstrapped bias on the R/L channel amps to max out V swing)

The Beta22 clearly can handle a larger range of cans but neither should have readily identifiable errors in their applicable ranges
Reply
post #10 of 24
3/2/10 at 12:32pm
Quote:
Originally Posted by jcx View Post
with the output biased Class A, having adequate I, Vswing for the can's Z there is some question of whether you might be able to hear differences - especially with controlled testing using level matching and blinding protocols

the M3 will be slightly noisier on the input than the discrete fets of the beta22, but attenuator Z, source resolution or recording noise floor could all be higher than either amp's input noise

the AD8610 is really quite recent, built on a very advanced process by a competent design team - its not likely to have major flaws or be comparable to 70's op amps that could have started the whole "discrete is always better" thing going

keeping in the vein of OPA627 substitutes there is the very new OPA827 from TI which would allow up to +/-18 V supplies for greater headroom with higher Z cans

I don't know if the M3 board layout and gnding could be improved, but the basic topology should be capable of excellent performance with reasonable load Z, sensitivity cans (I would change the cascode ccs to bootstrapped bias on the R/L channel amps to max out V swing)

The Beta22 clearly can handle a larger range of cans but neither should have readily identifiable errors in their applicable ranges
That's a very insightful post so I have to thank you for that, jcx.

Suppose that I use a PPAv2 now. Would the M^3 be a significant improvement or going straight to the B22 would be more worthwhile?
Reply
post #11 of 24
3/2/10 at 12:39pm
ppa series and m3 are very similar. I consider them peers, basically.

both are very quiet, very low distortion and for most people, they'd be more than enough to do the job.

its affordable so I'd start there. learn one (either one) and THEN move on IF you feel the need.

why jump right to something else without learning the middle steps? part of the fun is 'experience', isn't it?
Reply
post #12 of 24
3/2/10 at 1:46pm
Quote:
Originally Posted by linuxworks View Post
ppa series and m3 are very similar. I consider them peers, basically.

both are very quiet, very low distortion and for most people, they'd be more than enough to do the job.

its affordable so I'd start there. learn one (either one) and THEN move on IF you feel the need.

why jump right to something else without learning the middle steps? part of the fun is 'experience', isn't it?
I see... the M^3 is significantly cheaper especially more so since I already have a working σ11 and I plan to optimize the M^3 around the ADA4627-1 if I were to build one. I did a rough BOM and the total cost is about $150 minus casework and shipping.

But then again, if it sounds similar to a PPAv2 then I suppose going straight to a β22 will probably provide more improvement in the bigger picture.
Reply
post #13 of 24
3/2/10 at 2:09pm
Quote:
Originally Posted by K3cT View Post
But then again, if it sounds similar to a PPAv2 then I suppose going straight to a β22 will probably provide more improvement in the bigger picture.
I don't think it is necessarily a matter of the two sounding similar - you could certainly make a PPA and M3 sound very different based on OPAMP and output BJT choice - but rather that they occupy the same space in terms of build cost and key technical features.

If you are looking at something to advance your DIY skills, I would actually argue that the PPA is a more 'difficult' build than the M3. It has a higher part count, and a bit more wiggle room for various optimisations. The B22 is certainly a step up in difficulty, but also cost. I have no opinion on sound quality because I've never heard a B22.

So it really depends on what you want to get out of it. In a similar situation to yours, I decided I didn't want to build a B22 so I went completely sideways by building a Millett Max. I'm glad I did, because it was a really fun build, and a great sounding and looking amp for the price.
Reply
post #14 of 24
3/2/10 at 3:51pm
Thread Starter 
Wow, regardless of going a bit off topic of my question in the op you guys answered some other questions I was thinking of asking by going off track a little

The version of dt880 I have is 250ohms. The m3 should be enough to drive them from what I've read (correct me if I'm wrong).

Beefy, where do you get your supplies for amp building?
Reply
post #15 of 24
3/2/10 at 3:55pm
Quote:
Originally Posted by pwrusr View Post
The version of dt880 I have is 250ohms. The m3 should be enough to drive them from what I've read (correct me if I'm wrong).
Should be fine.

Quote:
Beefy, where do you get your supplies for amp building?
Mouser is always my first stop for components.
Reply
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Been gone for a while, is m3 still good for DIY?