Hello everyone,
My Tannoy Precision 8D dont work anymore so I opened them to check what went wrong.
I need to replace both crossover networks, plus one of the amplifiers - those parts are not available anymore from Tannoy.
Their design is based on two boards: "filter" board and "amplifier" board - or so they say. I found both of those cards inside the speaker, plus an "extra" card mentioned nowhere in the documentation.
I did a spectrum analysis of both outputs of the "filter" board and I traced the "extra" card schematic, screenshots attached.
I'm finding it hard to understand what kind of crossover design I need - I don't know much about speakers ; and the more research and tests I do, the more puzzled I am - so if anyone here was able to help it would be much appreciated 🙂
So far, I believe that the "filter" board is active crossovers because of the energy added to the signal as you can see on spectrum screenshots:
- dark blue is the noise fed into the board,
- light blue is the output fed to the amp of the bass driver,
- light green is the output fed to the amp of the tweeter (still full band at that stage).
This tweeter signal is fed, after amplification, into the "extra" board (passive crossovers I believe) to split it between the Tweeter and the Super Tweeter - which reproduces frequencies above 21 kHz according to documentation.
1) I don't understand if the frequency dip around 4 kHz on tweeters signal is intentional ?
2) I dont understand the values of the components on the schematic, I copied exactly what was written on them: can you help me decipher at all ?
3) I was told that L1 and R1 form a Zobel network, but after a bit of reading it is too complicated for my grasp - can anyone explain their function in layman's terms ?
4) There is only C1 in the tweeter signal path, which I understand to be a 1st order HighPass, hence -6 db/octave. But the crossover of the bass signal shows a much steeper filtering (about -32 db/octave or so if i'm not mistaken). Wouldn't it result in a funny overall frequency response of the speaker around cutoff ? Tannoy states that the crossover frequency is 2.2 kHz.
5) The monitors are built with Tannoy's "dual concentric" technology: the tweeter is located in the middle of the bass driver. Should the Lo-Freq and the Hi-Freq signals (post crossovers) be perfectly phase matched ? What about the Super Tweeter - which is not aligned with the other two drivers ?
6) I also see that frequencies are cut at -10 dB/octave under about 40 Hz on the bass signal. Would this be intentional on Tannoy's part? should I do it ?
Any input welcome, and thanks in advance to anyone who will look into it!
My Tannoy Precision 8D dont work anymore so I opened them to check what went wrong.
An externally hosted image should be here but it was not working when we last tested it.
I need to replace both crossover networks, plus one of the amplifiers - those parts are not available anymore from Tannoy.
Their design is based on two boards: "filter" board and "amplifier" board - or so they say. I found both of those cards inside the speaker, plus an "extra" card mentioned nowhere in the documentation.
I did a spectrum analysis of both outputs of the "filter" board and I traced the "extra" card schematic, screenshots attached.
I'm finding it hard to understand what kind of crossover design I need - I don't know much about speakers ; and the more research and tests I do, the more puzzled I am - so if anyone here was able to help it would be much appreciated 🙂
So far, I believe that the "filter" board is active crossovers because of the energy added to the signal as you can see on spectrum screenshots:
- dark blue is the noise fed into the board,
- light blue is the output fed to the amp of the bass driver,
- light green is the output fed to the amp of the tweeter (still full band at that stage).
This tweeter signal is fed, after amplification, into the "extra" board (passive crossovers I believe) to split it between the Tweeter and the Super Tweeter - which reproduces frequencies above 21 kHz according to documentation.
1) I don't understand if the frequency dip around 4 kHz on tweeters signal is intentional ?
2) I dont understand the values of the components on the schematic, I copied exactly what was written on them: can you help me decipher at all ?
3) I was told that L1 and R1 form a Zobel network, but after a bit of reading it is too complicated for my grasp - can anyone explain their function in layman's terms ?
4) There is only C1 in the tweeter signal path, which I understand to be a 1st order HighPass, hence -6 db/octave. But the crossover of the bass signal shows a much steeper filtering (about -32 db/octave or so if i'm not mistaken). Wouldn't it result in a funny overall frequency response of the speaker around cutoff ? Tannoy states that the crossover frequency is 2.2 kHz.
5) The monitors are built with Tannoy's "dual concentric" technology: the tweeter is located in the middle of the bass driver. Should the Lo-Freq and the Hi-Freq signals (post crossovers) be perfectly phase matched ? What about the Super Tweeter - which is not aligned with the other two drivers ?
6) I also see that frequencies are cut at -10 dB/octave under about 40 Hz on the bass signal. Would this be intentional on Tannoy's part? should I do it ?
Any input welcome, and thanks in advance to anyone who will look into it!
Attachments
Why do you need to replace the filter boards? Surely they are working OK as you obtained measurements from them. Although the lack of (active) high-pass filtering on the tweeter is a bit odd.
C1 is 8.2uF 5% tolerance 250V capacitor
C2 is 0.68uF 5% tolerance 250V capacitor
L1 is 0.026mH (26uH) inductor
R1 is 4.7R 5% tolerance resistor
C1 is 8.2uF 5% tolerance 250V capacitor
C2 is 0.68uF 5% tolerance 250V capacitor
L1 is 0.026mH (26uH) inductor
R1 is 4.7R 5% tolerance resistor
thanks for the component values Richie00boy.
Only one of the filter boards is still functioning, the other one is fried and no replacement boards are available from Tannoy or their distributors... I'm on my own here.
Only one of the filter boards is still functioning, the other one is fried and no replacement boards are available from Tannoy or their distributors... I'm on my own here.
I would expect the filters to be active versions.
I would also expect some EQ to have been applied either in the filter boards or in the inputs to the amplifier boards.
My Reveal 6D have 2 banks of dip switches but only two boards inside. It looks like one board is purely amplifier 2 channels for active operation, so I guess the other board does all the filtering + EQ + implementing the dip switches.
I would also expect some EQ to have been applied either in the filter boards or in the inputs to the amplifier boards.
My Reveal 6D have 2 banks of dip switches but only two boards inside. It looks like one board is purely amplifier 2 channels for active operation, so I guess the other board does all the filtering + EQ + implementing the dip switches.
yes Andrew, the filter board is equipped with a set of 20 dip switches that - I believe - implement control over the six MC33079P (datasheet attached) chips present on this same board.
I did the spectrum analysis with those switches positioned for a flat system response (see user manual screenshot attached).
Here I see the High-Pass filtering below 40 Hz so I assume it is intentional on Tanoy's part. Would anyone know why it could be required ?
I did the spectrum analysis with those switches positioned for a flat system response (see user manual screenshot attached).
Here I see the High-Pass filtering below 40 Hz so I assume it is intentional on Tanoy's part. Would anyone know why it could be required ?
Attachments
I forgot to mention that the drivers load is 8Ω for both tweeter and woofer. Regarding question 4, i calculated the cutoff frequency of the crossover applied to the signal sent to the tweeter according to formula f=1/(2*π*R*C)
With C1=8.2uF, I find a cutoff frequency of about 2.425 kHz. However, Tannoy states that the cutoff frequency of the crossovers is 2.2kHz so C1 should be about 9uF if I dont make any mistakes. Can anyone understand why? which value should i tune the cutoff frequency of my crossovers to ?
With C1=8.2uF, I find a cutoff frequency of about 2.425 kHz. However, Tannoy states that the cutoff frequency of the crossovers is 2.2kHz so C1 should be about 9uF if I dont make any mistakes. Can anyone understand why? which value should i tune the cutoff frequency of my crossovers to ?
It is a filter to cut off low bass, to allow the fairly small speaker to run less stressed. As it is a ported design, below tuning frequency it is wise to cut the sound.
Not familiar with the teeny Tannoys. only the DC varieties.
But typically the DC design is Two drivers mounted coaxially
Distinct/separate drivers, Each requiring it's own Crossover circuit
Are yours of entirely differing design then? Asking as there is No such circuit in your diagram.
Perhaps it's merely an adjunct to the LF cone/voice coil ? as DC's really do not need/benefit from an additional tweeter.
Just idle curiosity
But typically the DC design is Two drivers mounted coaxially
Distinct/separate drivers, Each requiring it's own Crossover circuit
Are yours of entirely differing design then? Asking as there is No such circuit in your diagram.
Perhaps it's merely an adjunct to the LF cone/voice coil ? as DC's really do not need/benefit from an additional tweeter.
Just idle curiosity
I had a look at the DCs, they are what Tannoy calls "dual concentric" meaning the tweeter is mounted directly in the middle of the woofer cone. Mine are the same: yes, distinct drivers requiring separate crossover units.
Here is a quote from a precision range brochure where dual concentric technology is described :
"the Dual Concentric™ driver is actually two drivers properly merged into one. The high-frequency driver is positioned on the back of the low frequency magnet and fires through the centre of the cone. The result is that the acoustic apex of the high and low frequency drivers is aligned in three axes; a true point source."
http://www.tannoystudio.com/media/1146/precision-range.pdf
And then, above, is located the Super Tweeter (look at the photo in my initial post on this thread, should give you more insight)
Here is a quote from a precision range brochure where dual concentric technology is described :
"the Dual Concentric™ driver is actually two drivers properly merged into one. The high-frequency driver is positioned on the back of the low frequency magnet and fires through the centre of the cone. The result is that the acoustic apex of the high and low frequency drivers is aligned in three axes; a true point source."
http://www.tannoystudio.com/media/1146/precision-range.pdf
And then, above, is located the Super Tweeter (look at the photo in my initial post on this thread, should give you more insight)
Sorry to dig up such an old thread, but that passive crossover is only for the coax tweeter that's "inside" the woofer, and the supertweeter that's external, near the top of the enclosure.
Unless that brown glue totally dried out and went uber corrosive, you shouldn't need to replace the entire filter boards.
In these speakers (Reveal and Precision 6D and 8D), the two 5W ceramic dropper resistors ahead of the 7815/7915 regulators often fail open-circuit, so the opamps on the filter board no longer receive power. Also, the switches can go wonky as well.
You can find the schematics here:
http://91.121.194.115/tannoyschematics.zip
The filter board is almost definitely the same. The only (electronic) difference between the 6D and the 8D is that the 8D has two bridged LM3886's driving the woofer, where the 6D only has a single one.
Unless that brown glue totally dried out and went uber corrosive, you shouldn't need to replace the entire filter boards.
In these speakers (Reveal and Precision 6D and 8D), the two 5W ceramic dropper resistors ahead of the 7815/7915 regulators often fail open-circuit, so the opamps on the filter board no longer receive power. Also, the switches can go wonky as well.
You can find the schematics here:
http://91.121.194.115/tannoyschematics.zip
The filter board is almost definitely the same. The only (electronic) difference between the 6D and the 8D is that the 8D has two bridged LM3886's driving the woofer, where the 6D only has a single one.
>Although the lack of (active) high-pass filtering on the tweeter is a bit odd.
Looks like I’ve understood the entire filtering scheme.
I initially expected that a 2 way (actually 3 way with an extra wide-band tweeter took into account) has to have HiPass and LowPass filters connected to corresponding amplifier and finally to speakers. Such approach was implemented in any 2 way system I saw, but not in Tannoy Precision (6D in my case)
In the 6D there actually is a LowPass active filter connected to the first amplifier and to the woofer,
But all that could be treated as HiPass filter actually has the flat response (I’ve measured) , so HiPass filtering is made by the passive filter that pierrepierrepierre called “crossover”.
Formally it’s a crossover as well. It separates frequencies between the tweeter and the wide-band, but it’s also a HiPass passive filter for the both.
Why so? My humble hypothesis is the equalization possibilities of the 6D.
As the equalization happens in the mid frequencies as well, for “true” 2-way active crossover we wound need some synchronized equalization circuits in the both channels as equalization affects the range of the crossover frequency, But it would be too complex electrically.
So the vendor just decides to move HiPass filtering to the passive part.
Maybe there is another reason for such approach, but I could hardly imagine a simple circuit that is able to equalize the response equally in the both (Hi+Low) channels.
How did I detect that?
I have a pair of 6Ds and found its sound is not as good as it should be.
The sound was obviously good at the time the 6Ds were developed, but not now.
There was a lot of distortion practically measure (and audible) at the existing amplifier output.
So I’ve nearly completely redesigned the active part of the 6D.
The filtering trick explained above forced me to keep the crossover board so far,
but I removed the transformer and the amplifier+power_supply board and installed
IcePower 125ASX2 (class D) and +/-15v voltage regulator (required to power the crossover/filter board).
I see many people could think that classD is not acceptable for high quality design.
I thought the same before I faced the IcePower.
If somebody would like to know more I would be happy.
My next step is replacing the crossover/filter board with true crossover as well and removing the HiPassing abilities from the passive filter.
Looks like I’ve understood the entire filtering scheme.
I initially expected that a 2 way (actually 3 way with an extra wide-band tweeter took into account) has to have HiPass and LowPass filters connected to corresponding amplifier and finally to speakers. Such approach was implemented in any 2 way system I saw, but not in Tannoy Precision (6D in my case)
In the 6D there actually is a LowPass active filter connected to the first amplifier and to the woofer,
But all that could be treated as HiPass filter actually has the flat response (I’ve measured) , so HiPass filtering is made by the passive filter that pierrepierrepierre called “crossover”.
Formally it’s a crossover as well. It separates frequencies between the tweeter and the wide-band, but it’s also a HiPass passive filter for the both.
Why so? My humble hypothesis is the equalization possibilities of the 6D.
As the equalization happens in the mid frequencies as well, for “true” 2-way active crossover we wound need some synchronized equalization circuits in the both channels as equalization affects the range of the crossover frequency, But it would be too complex electrically.
So the vendor just decides to move HiPass filtering to the passive part.
Maybe there is another reason for such approach, but I could hardly imagine a simple circuit that is able to equalize the response equally in the both (Hi+Low) channels.
How did I detect that?
I have a pair of 6Ds and found its sound is not as good as it should be.
The sound was obviously good at the time the 6Ds were developed, but not now.
There was a lot of distortion practically measure (and audible) at the existing amplifier output.
So I’ve nearly completely redesigned the active part of the 6D.
The filtering trick explained above forced me to keep the crossover board so far,
but I removed the transformer and the amplifier+power_supply board and installed
IcePower 125ASX2 (class D) and +/-15v voltage regulator (required to power the crossover/filter board).
I see many people could think that classD is not acceptable for high quality design.
I thought the same before I faced the IcePower.
If somebody would like to know more I would be happy.
My next step is replacing the crossover/filter board with true crossover as well and removing the HiPassing abilities from the passive filter.