The POOR man's Tannoy DMT - AKA "The PM-DMT"

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
It is remarkable what you can find on the WWW with a bit of patience. I'm scanning the WWW for years and bought my complete hybrid studio (run digital audio through an AD/DA convertor to an analog mixing desk, to use the character of the analog desk) all second hand for around 1/5 to 1/10 of the new price. ALL in perfect shape!
 
I did ask the question, because it stills intrigues me. Is a middle of the road driver (CHEAP) combined with a correct filter (correcting to make it within +- 2 dB) sounding the same as a brilliant (EXPENSIVE) driver that is +- 2dB BY ITSELF???

This topic started in my head when I saw the Danley Synergy horn (commercial product) using VERY cheap mid drivers, most likely because they just matched exactly with what was needed.

Depends on the needs of the app, though the more drivers used to break up the BW, the less it matters. For instance, some decades ago a $80k+ system used a wide range mid-bass that when its label was removed proved to be a European driver I could buy at a Doraville [small town in N.E. Atlanta] DIY audio emporium for ~$13 w/tax, so inflation adjusted, ~$20 before any rate changes/VAT/whatever, but the XO was a doozy [big/complex] using the best components to rave reviews.

Re the Synergy concept horns; the drivers are used in BP4 and or BP6 alignments at relatively low power, so its T/S performance specs, construction type, unit cost to meet the needs of the app are the only criteria. Based on miscellaneous bits n' pieces of info posted around the net, it appears they were designed specifically for the DSL products, hence the 'perfect' set of specs.

GM
 
For 2 identical drivers in parallel that would make;

Vas = 2x driver Vas
Sd = 2x driver Sd
Re = driver Re / 2
Lvc = 2x driver Lvc
Pe = 2x driver Pe
SPL = SPL of single driver + 6dB

Mechanical;
Mms = 2x driver Mms


I would expect the following T&S parameters to stay the same;
Fs, Qts, Qms, Qes, BI and Xmax.

For a PARALLEL connected second driver, the TL will just double in width.

Close! ;)

Lvc = Lvc/2

Driver conversions

GM
 
Tannoy seems to have had different versions of their 8 inch drivers.

I found that ~ 1.8KHz was achievable with a third order slope.

Seems reasonable, an 8" is 90 deg/~1,692 Hz, so not that far off plus it s/b 3rd order since the woofer has a 90 deg delay whereas the compression horn has none, though normally it's an acoustic 3rd order when the HF driver is behind the woofer.

GM
 
Seems reasonable, an 8" is 90 deg/~1,692 Hz, so not that far off plus it s/b 3rd order since the woofer has a 90 deg delay whereas the compression horn has none, though normally it's an acoustic 3rd order when the HF driver is behind the woofer.

GM

This Tannoy 2036, which is an 8" Coaxial, seems to be crossed @ 1.7 kHz in the 8 PMT II, which makes it all fit.

So, for the 3-way Main Monitor we try to build here, the Tannoy will be High passed @ 300 Hz and Low passed @ 1.7 kHz.

kind regards,

Frans
 
I would have thought the main difference to be Fs.

8NDL51 with 66Hz and the 8FG51 with 49Hz.
Being a good third of an octave that should be noticeable.

I started with the 8NDL51 to sim it, than used the same TL to put in the 8FG51 which worked as a drop in replacement. Being kind of lazy, but that's not the best way to do it. I did a quick redesign today, the 8FG51 digs deeper, at the cost of a longer line of course.

Will do a complete recalc for both drivers, using the standard approach as also described by King. Adding the way how I came to the TL clearly described. That way, I will learn the TL theoretical back ground. Also did find the "Accidental MLTL technique by XRK971.

Looks like the 8NDL51 will win @ Spl, by about 1.5 dB. Will try to find the time this week.

kind regards,
Frans
 
The resonance frequency of the tweeter is a factor for the crossover. Input to the tweeter has to be well down at this point if the speaker is going to be driven hard. This is why I ended up with ~1.8KHz. Measurements picked up the resonance with the crossover in place. A DSP might be able to notch it out, which is preferable to steeper filters.
 
The resonance frequency of the tweeter is a factor for the crossover. Input to the tweeter has to be well down at this point if the speaker is going to be driven hard. This is why I ended up with ~1.8KHz. Measurements picked up the resonance with the crossover in place. A DSP might be able to notch it out, which is preferable to steeper filters.

Hi spladski,

Did you measure that on the Tannoy 2036, or a different driver. Do you still have those measurements and are you willing to share? I do intend to use the DSP in the Hypex 3-way module.

kind regards,
Frans
 
Hi spladski,

Did you measure that on the Tannoy 2036, or a different driver. Do you still have those measurements and are you willing to share?
Frans
I don't have the individual measurements to hand. I managed to dig out the final measurement. You can clearly see the resonance at 900Hz. This was a passive xover and very hard work to achieve this level of performance. The microphone I was using was a button electret which couldn't be trusted entirely. Note the absence of any mid band suckout which can plague coaxials. Disregarding the one peak, it came out at +/- 1.5dB for most of the important mid range. Now I use B&K 4165 mic as I found that once a speaker gets to within +/-2dB, 0.5dB trims can be heard and need to be measured.

The peak does not manifest itself in a bad way. It actually makes drums sound more dynamic than they really are. No speaker is perfect and I have encountered some truly dreadful 'studio' class monitors.
 

Attachments

  • Tannoy_Saturn.PNG
    Tannoy_Saturn.PNG
    15.1 KB · Views: 240
As a follow up, I dug out some more info. The brochure for the NFM8 indicates 2.3KHz for the crossover, but the driver is listed as the 2025. It also states low order crossovers so this would explain the higher frequency.

Another point is the frequency response. The one I posted may not look as flat as people claim for other designs. It is a single measurement, not averaged. What it does indicate, is that with DSP +/- 1dB maybe possible.

In another thread someone asks what should be the design goal. Not all measurement methods produce the same results which complicates matters. The Saturn had a review with a frequency response plot. This showed mild midrange depression. However, when I measured it, the mid range depression was severe. So either my method differed from MLSSA which is commonly used, or Tannoy had a different crossover revision.
I don't have access to MLSSA, so cannot compare directly. However, it seems to me that MLSSA does not reveal all the gory details because it is an indirect method. I use gated sine wave which reveals edge diffraction effects from the drivers themselves, not only the cabinets. I also make measurements at less than 1 meter which accentuates defects.
 

Attachments

  • NFM8.PNG
    NFM8.PNG
    32.6 KB · Views: 236
  • Saturn S8LR.jpg
    Saturn S8LR.jpg
    34.7 KB · Views: 211
Last edited:
As a follow up, I dug out some more info. The brochure for the NFM8 indicates 2.3KHz for the crossover, but the driver is listed as the 2025. It also states low order crossovers so this would explain the higher frequency.

Correct, how-ever I have the Series II, which has a 2036 Driver and passive filter Second Order Low, First Order High @ 1700 Hz. There is a spreadsheet at the Tannoy Hilberink site I got the info from, which is consistent with my series II brochure.

Must be quit a different driver, also higher Pe (from 120 to 200 Watt).

The lowest cross-over I found, Tannoy used on DC's is 1000 Hz, highest 2500 Hz (on the the 2033 & 2041). They go all the way up from 1000, 1100, 1200 etc.


Another point is the frequency response. The one I posted may not look as flat as people claim for other designs. It is a single measurement, not averaged. What it does indicate, is that with DSP +/- 1dB maybe possible.

Which is why a opted for only DSP versions. As a side note, maybe just as important, I don't have extensive knowledge on passive cross-overs just understand the basic principles. :eek:

In another thread someone asks what should be the design goal.

These should become my "Studio Main monitors" for the project studio market. Meaning neutral response at high SPL levels (relative to the smaller control-room). VERY important conditions; LOW cost components that are commercially available, the speaker should come very close to "Top notch" like ATC, PMC and the likes, NO sub required.

Not all measurement methods produce the same results which complicates matters. The Saturn had a review with a frequency response plot. This showed mild midrange depression. However, when I measured it, the mid range depression was severe. So either my method differed from MLSSA which is commonly used, or Tannoy had a different crossover revision.
I don't have access to MLSSA, so cannot compare directly. However, it seems to me that MLSSA does not reveal all the gory details because it is an indirect method. I use gated sine wave which reveals edge diffraction effects from the drivers themselves, not only the cabinets. I also make measurements at less than 1 meter which accentuates defects.

I will get some assistance in measuring, we will state the method used.

Kind regards,

Frans
 
Hi ! another approach is the one by Geithain speakers

166022.jpg


going back to Tannoy i would think to a satellite plus sub solution, with the satellite made out of a DC and a very nice 15" below. The choice is endless.
Problem is which DC ? there are many old DC on the used market that could be used with nice results i guess.
I have a pair of DC100 and the drivers look very very serious. I bought also a pair of DC1000 broken and the drivers were less nice I did not like them
I have them stored away and i will try to see if they are still working after all these years.

and then which size ? ... a 6" or a 8" ? and cut at which frequency in the bass ? I would use a 8" cut at around 120Hz ...
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.