Jump to content

Sign in to follow this  
ThirdDrawerDown

Another Maggie Tympani-1D rebuild thread

Recommended Posts

49 minutes ago, gonefishing999 said:

Hmm. Maybe get @andyr to have listen to a Pro. Amp. In all it's class AB glory.:emot-bang:

 

My current amps are all Class AB, Peter.  So I know how they sound.  They sound as good as they do because of the way Hugh Dean organised their harmonic profile.  🙂

 

Quote

By the way Andy I also have the a Lexicon CX-5 .

 

Yes, you said a few posts ago.

 

Andy

 

Edited by andyr

Share this post


Link to post
Share on other sites
4 hours ago, gonefishing999 said:

I haven't forgot you Murray

I have a heavy bout of sanding to do on the plasterboard in the listening room, and an overdose of business travel and organisational change coming up, so for all my woes I remain in no hurry!

 

Green glue is my friend.

Share this post


Link to post
Share on other sites

Some frequency sweeps.


XO points are as follows: 80 Hz, 1100 Hz, 3000 Hz

 

6db BW

20180611_183122.jpg

Share this post


Link to post
Share on other sites




Bessel.

Just to add NO PEQ's or FIR's. 

Did some basic leveling.


Obviously drop out at 100 Hz.

 

This is a minisharc - not minidsp.

20180611_185004.jpg

Share this post


Link to post
Share on other sites

@ghost4man  Maggie's work very well in your room not much EQ to fix up.

Are the graph of one speaker? Is the other close to this one.

Looks like out of phase at 80Hz xo and TR need 4-5 db increase.

 

Share this post


Link to post
Share on other sites
On 11/06/2018 at 7:06 PM, ghost4man said:

Some frequency sweeps.


XO points are as follows: 80 Hz, 1100 Hz, 3000 Hz

 

6db BW

20180611_183122.jpg

 

Interesting, Ozzie - but I don't understand what you are saying with this graph:

  1. you say "XO points are as follows: 80 Hz, 1100 Hz, 3000 Hz" and "6db BW"
  2. do you mean all HP & LP filters are @ 6dB?
  3. this is for your T-1Ds.  Can you be a bit more precise - IOW:
  • bass LP is 6dB @ what ... 80Hz?  (If so - extraordinarily low?)
  • old tweeter/new mid HP is what ... 1100Hz?
  • old tweeter/new mid LP is 1100Hz?
  • and TR HP is 3000Hz?

I'd like to know what the @ frequencies are for each XO between different drivers - just bcoz they're 6dB slopes doesn't mean the LP & HP '@ frequencies' have to be the same.

 

Don't know what the black line represents ... but your overall FR graph (the upper purple line) is pretty level but is up compared to <400Hz - whereas I thought the ideal was a gradually sloping downwards line with increasing frequency?

 

Andy

 

Share this post


Link to post
Share on other sites


Andy, 

 

Just to be clear this was something of first up test to get the ball rolling and see what results would throw up. 

 

The whole point was to see if there was much in the way noticeable crests or troughs which I experienced before with lesser amplifiers. 

 

In answer to your question I left the HP and LP with the same slope. So for example, when I tested with 12dB LR I kept that consistent across the board. 

 

I am well aware of Magnepans recommendations. Magnepan however don't build a system with a DSP unit. 

 

Will I test with standard Magnepan metrics? Well yes. However, I will ultimately rely upon what my ears tell me. 

 

The bottom end is catered for by a pair of 12 inche Dayton Audio drivers which I cut off at about 80Hz. The two bass panels kick in and go through to 1100Hz where they are cut off. It's at this point I have introduced the tweeter panel which I run to 3000Hz and cut off at this point to bring in the true ribbon which is basically a super tweeter. 

 

I am going to experiment with the TR from 2k through to 10k to see what graphs are yielded and more importantly how it sounds. 

 

A gradually falling slope is the preferred option but im this case it's not something that I am overly concerned about. My feeling is a 6 dB slope down is a good guide. 

 

The drop off at 100 Hz is not something I am happy with. 

 

This is the beauty of a DSP unit where you can build your own XO to test. Overall I'm very positive about the performance of the Lexicons. Far superior to the AKSA's. 

 

Out of curiosity I am going to run the old tweeter full range. My guess is that it will fall over despite Magnepans assertions. 

 

What HP, LP markers would you suggest? I'm open to suggestions. Enjoying the process of testing again. 

Share this post


Link to post
Share on other sites
On 13/06/2018 at 10:07 PM, andyr said:

do you mean all HP & LP filters are @ 6dB?

Andy if you look at top left corner of each graph Ozzie has put db - BW or LR  .

If you compare each graph there is not much difference all graphs so for me 48 db LR rocks.

Ozzie said he preferred 48 db LR so much for your CAD XO simulations .

Edited by gonefishing999

Share this post


Link to post
Share on other sites

I think he’s gone to Russia for the World Cup.

Share this post


Link to post
Share on other sites


On 14/06/2018 at 7:50 PM, ghost4man said:

Andy,

 

Just to be clear this was something of first up test to get the ball rolling and see what results would throw up.

 

The whole point was to see if there was much in the way noticeable crests or troughs which I experienced before with lesser amplifiers. 

 

In answer to your question I left the HP and LP with the same slope. So for example, when I tested with 12dB LR I kept that consistent across the board.

 

Understood, Ozzie - and now I understand what your XO frequency numbers mean.

 

I am well aware of Magnepans recommendations. Magnepan however don't build a system with a DSP unit.

 

They certainly don't sell a system with a DSP unit - but how do you know they don't use one in their design process for a new model - before they actually release it as a finished product, with the passive filters they chose?

 

Will I test with standard Magnepan metrics? Well yes. However, I will ultimately rely upon what my ears tell me.

 

Absolutely!  🙂

 

The bottom end is catered for by a pair of 12 inch Dayton Audio drivers which I cut off at about 80Hz. The two bass panels kick in and go through to 1100Hz where they are cut off. It's at this point I have introduced the tweeter panel which I run to 3000Hz and cut off at this point to bring in the true ribbon which is basically a super tweeter.

 

Aaahhh - understood!  Except my definition of what you are doing is that you are using the (T-1D) tweeter panel as a mid - from 1100Hz to 3000Hz - and then bringing in the TR as the tweeter.  (Super tweeters tend to start upwards of 10kHz!)

 

I am going to experiment with the TR from 2k through to 10k to see what graphs are yielded and more importantly how it sounds.

 

Absolutely the sound is the key, here.  I would suggest only using a roll-off point down at 2kHz for the TR if you use 24dB and 48dB slopes - otherwise you will be straining it.  Blowing TRs is an expensive exercise!

 

A gradually falling slope is the preferred option but im this case it's not something that I am overly concerned about. My feeling is a 6 dB slope down is a good guide.

 

Agreed - so I'm surprised Peter is telling you to increase the TR level by 4-5dB??

 

The drop off at 100 Hz is not something I am happy with.

 

Could this be the result of the distance the bass panels are off your front wall?  Try changing this distance and see if the 100Hz changes to something else.

 

This is the beauty of a DSP unit where you can build your own XO to test. Overall I'm very positive about the performance of the Lexicons. Far superior to the AKSA's. 

 

Out of curiosity I am going to run the old tweeter full range. My guess is that it will fall over despite Magnepans assertions.

 

Sorry, don't understand what you mean?  in a stock T-1D ... the tweeter panel does run full range - well, from 1100 to 20kHz, anyway!  But if you mean you won't use a LP filter for the new mid panel, to match where the TRs come in ... easily tryable with the DSP unit - but it means that you have 2 drivers operating together, from 3kHz (or whatever you choose) to 20kHz.  I suggest this will muddy the sound - compared to rolling off the new mid panel at the frequency the TRs come in at.

 

 

Comments above.

 

On 15/06/2018 at 8:37 AM, gonefishing999 said:

Andy if you look at top left corner of each graph Ozzie has put db - BW or LR.

 

Thanks, Peter - I had missed that.

 

On 15/06/2018 at 8:37 AM, gonefishing999 said:

Ozzie said he preferred 48 db LR ...

 

I presume you mean, Peter, that he prefers listening to the 48dB L-R?  That's great but almost no-one on the Planar Asylum uses steep slopes like that.

 

Not that that matters!  What matters is that Ozzie likes the sound he gets from 48dB L-R - so who cares what everyone else thinks!

 

He has inspired me to play around with different slopes on my Frankenpans.  👍

 

On 15/06/2018 at 8:37 AM, gonefishing999 said:

so much for your CAD XO simulations.

 

You don't use XO simulation programs to compare different L-R slopes against each other - you use them to see, for instance:

  • where the roll-off points (the '@ frequencies') need to be, to get the smoothest response when you have different slopes for HP & LP.
  • how different component values in a passive XO change the filter response.

 

Andy

Share this post


Link to post
Share on other sites

@djb you didn't sight me because I was out of sight due to being cited as I reacted, not responded, to a lack of citation.

 

[thread resumes]

 

Over the past week my stock of valves has reduced again due to a couple more of the KT88 tubes reaching their life's end. The time for the EL34 experiment may have arrived; I'm glad I bought an octet here on SNA last month (thanks again to the seller). I will be using them in a Cary 120S and I'm sure it will be straightforward, the only caveat is here:

 

https://www.upscaleaudio.com/products/electro-harmonix-el34

 

Quote
sheldon simon on Jun 22, 2012

I called cary audio to confirm that EL34 tubes were compatible with my cary 120s amp-even though the owners manual says they are.I spoke to a tech at Cary and they said sure-it will work perfectly-just drop them in. Guess what--not the case--the bias resist has to be modified to allow proper biasing to approx 140---without this mod won't get past 70ma----im confused as to why the amp manufacturer gave me incorrect info--but Kevin at upscale explained why they need adj and as usual he was correct.

Time will tell.  Sooner (if I need to use the EL34s immediately) or later (if my remaining KT88s are good for a while).

 

 

Edited by ThirdDrawerDown

Share this post


Link to post
Share on other sites
On 16/06/2018 at 12:45 PM, andyr said:

You don't use XO simulation programs to compare different L-R slopes against each other - you use them to see, for instance:

I understand what XO simulation apps do. Like I have said before what you are play around with

is only a simulation and should not be relied upon as accurate.

The microphone is the best method of testing different XO points. 

Share this post


Link to post
Share on other sites
30 minutes ago, gonefishing999 said:

 

The microphone is the best method of testing different XO points.

 

 

True, Peter - but you don't seem to have comprehended what I said - viz:

 

you use them to see, for instance:

  • where the roll-off points (the '@ frequencies') need to be, to get the smoothest response when you have different slopes for HP & LP.
  • how different component values in a passive XO change the filter response.

 

How do you do this with a mic and REW?  The way I see it ... you do the sim first  - because you can do this on a computer rather than having to buy many different parts and solder them in - then you measure your choice of sim with REW.

 

Andy

 

 

Share this post


Link to post
Share on other sites
1 hour ago, gonefishing999 said:

@ghost4man Ozzie vibered this to me last night . It is 2 way bass panels and mid/tweeter.

@andyr Andy you have stated on here that the the tweeter goes to 20khz .

 

IMG_7624.JPG

 

That's a very interesting FR, Peter.  It shows far less bass than mids & treble - which AIUI is not the desired profile (which should slope down from bass to treble)?  You have the REW graphs from when you measured David's T-1Ds - why don't you compare them with Ozzie's?  As I remember it, David's showed much more bass?  So why are Ozzie's T-1Ds so lacking in bass?

 

And re. "you have stated on here that the tweeter goes to 20khz" - I'm pretty sure what I actually wrote was "the tweeter is specced to 20khz".  In effect the same thing - it should certainly not have a roll-off starting at 9khz!  So I'm wondering why Ozzie's tweeters are rolling off so early.  Could it be the amp he is using?  :)

 

I can't remember - how does David's FR look like, in the HF region (driven by his Amber amp)?

 

Andy

 

Share this post


Link to post
Share on other sites

In relation to the above the following needs to be stated. 

 

The 2 graphs represent a two and three way set up. 

 

The graph with the obvious drop off at around 10k is where I have run the original tweeter from 1100 to 20k.

 

To my mind it's a very poor result. The T1Ds are specced to 20k and it certainly achieves that goal but not what I consider the ideal outcome. 

 

The other graph is three way and is defined by the TR being introduced at 3k with 48dB slopes. 

 

Clearly there is a significant difference between the graphs. 

 

If you draw the conclusion that the amplifier is struggling with the load then to my mind you are not correctly interpreting the results especially given the fact that the amplifier is dealing with a less than 3 ohm load when sitting on the TR. 

 

If the frequency response was reversed then the argument would conceivably apply. But not with the above. 

 

It simply shows that the TR is a far better performer in this range than the existing tweeter. 

 

Should we be surprised with the result? 

 

Absolutely not. It simply reinforces why Magnepan went down the route of adding the TR. 

Share this post


Link to post
Share on other sites
1 hour ago, ghost4man said:

To my mind it's a very poor result. The T1Ds are specced to 20k and it certainly achieves that goal but not what I consider the ideal outcome.

 

Something specced to 20kHz should be (at most) -3dB down at 20kHz.  Any more and it's not "delivering" to 20kHz.

 

Quote

If you draw the conclusion that the amplifier is struggling with the load then to my mind you are not correctly interpreting the results especially given the fact that the amplifier is dealing with a less than 3 ohm load when sitting on the TR.

 

The only reason I mentioned the amp, Ozzie, was because in a previous post (possibly not on Murray's thread) you said the reason why your REW plot showed the TRs falling off from 16kHz must be because of the amp you were using.

 

That's why I suggested Peter compare your 2-way T-1D plot against the one he made at David's place (before he added the ribbon tweeters).

 

Andy

 

Edited by andyr

Share this post


Link to post
Share on other sites
1 hour ago, andyr said:

Something specced to 20kHz should be (at most) -3dB down at 20kHz.  Any more and it's not "delivering" to 20kHz. 

That would only apply to an anechoic far field measurement on axis. With an in-room measurement we expect to see the response fall towards the top end - in fact that's also generally desirable. Even so, typically you will notice the difference between a response that is sloping down and one which is rolling off rapidly.

 

 

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...