WOPL Tri Amp ?

[Mohawk]

I want to put together a Tri amp system with 3 WOPL'ed amps... A 700B & 2 400's .... 15" JBL drivers in forward ported & tuned 177L cabinets , JBL 2445 Compression drivers / Horns and JBL 2405 Tweeters .....
I plan to feed the beasts from a Minidsp 4x10HD ..... Thinking 500Hz cut off for the 15's leaving 500+ to 7,000 for the horns and beyond 7k the tweeters can have the rest.
The compression drivers & tweeters are ultra efficient compared to a "cone " driver so, what would be a reasonable way to reduce the gain with the 400's ?
possibly an RCA attenuator feeding the 400's? (beyond what the DSP does) I'm looking to reduce the "idle" or no signal hiss from the horns/ tweets without muting the frequency response .... ???
Also , Joe's new design 400 light boards have jumpers that when set can provide a "sensitivity" adjustment...
Any idea on the resistor / cap values to display 20- 50 ish watts ginen the application ?

This make sense ?

M

 

[mlucitt]

I like your plans, however 500Hz is at the low edge of the 2445J's bandwidth. The WOPL'd PL400 is a serious machine in that frequency range. The 2445J's can handle 100W and 150W on a good day (above 1KHz). Now, I have firsthand knowledge that 100W going into a 2445J will make your ears bleed, that is not justification to say, I won't listen to it that loud. (I cranked my 2426J horns up all the way up and had to stand out in the garage until the thermoswitches in the amplifier shut it down, but that is another story)
I would recommend a crossover at 800Hz or even 1KHz, because the 2445J's are good to 20KHz. The upper crossover could also benefit from an increase to about 10KHZ or 12KHZ, but that is about the limit of my hearing threshold anyway.

 

[laatsch55]

And the slope is another criterion that deserves serious consideration. My extreme slope K's crosses are 96db/octave....

 

[mlucitt]

And the slope is another criterion that deserves serious consideration. My extreme slope K's crosses are 96db/octave....

I never have understood the slopes, but I take it the frequency is in a hurry to cutoff/cuton if the slope is steep? And if the slope is less steep the frequency is a little more casual about getting to the cutoff/cuton?

So, yeah, what Lee said; if your crossovers are serious, then your octaves are closer together and those WOPL'd decibels are going to pile up quickly on the backs of those horns and tweeters.

 

[laatsch55]

The slope Mark is when the cutoff frequency is reached, how fast is the gain dropped as the frequency rises from there..Less energy in the range not tolerated by the driver. The steeper, the better. VERY noticeable difference between the 6db to the 96db, like putting up a brick wall...

 

[Mohawk]

This great stuff !

Thank you for jumping in..... I like what Laatsch said about moving the crossover points up higher , Off the top of my head I think the Minidsp can go to 48 db/ octave ? I'll have to look that one up....
Thoughts on what I can do to lower the idle hiss ? Maybe L pad type of circuit ?
Check out this site I wandered into.

http://www.erseaudio.com/CrossoverCalculators/L-Pad-Attenuation



Would this concept work for clearing up the "hiss" while not affecting frequency response
set the calculator for -5 db on a 16 ohm driver and we get a 20ohm parallel and a 7ohm series .....

M

 

[wattsabundant]

Doesn't the Minidsp handle the crossover points and levels? If so, all that would be needed would be some coupling caps on the high frequency drivers for protection from potential DC offset. Also, depending on what 15" is used, 500hz may be pushing it. Is a 4 way system out of the question? The minindsp could handle it.

This does not appear to be a typical home system. Is it sound reinforcement?

 

[Mohawk]

Yes, the DSP will handle all the crossover points and levels.
What I'm looking to reduce is the no signal noise or "hiss" inherent with a super efficient compression drive and horn ....
I do have the dc protection caps ready
But like you say their only job is to protect the driver from DC or a lower frequency that it can tolerate
This is a system destin for the living room....
My wife has also questioned my sanity !

M

 

[laatsch55]

This great stuff !

Thank you for jumping in..... I like what Laatsch said about moving the crossover points up higher , Off the top of my head I think the Minidsp can go to 48 db/ octave ? I'll have to look that one up....
Thoughts on what I can do to lower the idle hiss ? Maybe L pad type of circuit ?
Check out this site I wandered into.

http://www.erseaudio.com/CrossoverCalculators/L-Pad-Attenuation

View attachment 41025

Would this concept work for clearing up the "hiss" while not affecting frequency response
set the calculator for -5 db on a 16 ohm driver and we get a 20ohm parallel and a 7ohm series .....

M

I didn't say anything about moving the points up higher, just making the crossover point like a brick wall to the unwanted frequency. IO think you should try the WOPL first before you worry about any hiss.....REAL HARD to hear in my K's....

 

[Mohawk]

Correct ,it was mluciit the suggested moving the crossover points up... Anyhow,
I have the recently repaired a full WOPL 700B The one with "more output on the left channel" ...
After a rigorous run on the bench it came back upstairs to power Paradigm Signature S8's

Beautiful rich and silent.... This is using the passive crossovers within the S8's
I now have most of the tri amp stuff ready, just one more stock 400 to WOPL ....

As I'm tinkering with this new project I set the dsp for crossover points and do some basic configuration

WOPL 700B crossed @ 500hz 177L cabinet JBL 2035's
WOPL 400 crossed @ 7,000 JBL 2445's / 2390 horn + wave guide
Stock 400 crossed @ 7,000 JBL 2405's

This was my starting point... The DSP allows me to jump all over the place in real time with levels ,crossover points !

The DSP is driving the JBL's directly except for the addition of protection caps in series with the 2405's & the 2445's
No protect caps required for the LF 2035's
The selected protect caps won't have any effect on the set frequency response.....

The 2445's are 111 db efficient. The 2035's by contrast are 95 db efficient , now add the 2390 horn to the 2445's
and we effectively have more amplification (horn effect ?).....

So, the same amp will make more audible "idle" noise or hiss with no actual input signal when driving the 2 differently
efficient speakers ? Sound right ???

Any of the 3 amps I have ready experienced the same effect.... More idle noise with the more efficient driver.....
But, all silent with the Signature S8's and their passive crossover networks....

So, my thought was to use the above "l pad " idea to replicate what the passive crossover did without messing up
frequency response ?


Interesting read on the idea....


https://www.diyaudio.com/forums/multi-way/252713-cheap-free-improve-horn.html



I think I need a Sunday morning beer.....

M

 

[Michael F]

I agree that a 500hz crossover point is pushing the 2445 a bit, does the horn its coupled to even support 500hz? You can safely bump up to 1khz where the 15" woofer will begin to beam. A Linkwitz Reilly fourth order filter (24db/8ve) would be a good starting point. Hiss is usually a product of a gain structure mis-match. Does your hiss vary with different level settings on the DSP? In either case an L-Pad wont help you as it attenuates indiscriminately, hiss and program material.
Be aware of JBL polarity conventions too. Once set up, pay particular attention to the frequency response at the crossover points. You may need to invert a band to obtain smoothest response.

 

[Mohawk]

Thank you for jumping in Michael,
I actually looked right past the 2390 specs and indeed 800hz is the crossover point for the 2390 horn . Thanks for pointing that one out !

The "hiss" seems to be a no signal "idle" situation.... The DSP settings have no effect without any output.... "Idle".
The 1st experiment was to connect the low efficiency 15" driver (no input) ... Negligible noise ... Same amp with the compression driver and we get the hissing.. Now to be clear this hiss is not driving me out of the room , just audible enough for me to want to kill it ....
As we get playing a track it vanishes into the music...
2nd test With the paradigm S8"s this hiss is a non issue... I credit the passive crossovers...
Again, just the protection caps in series with the 2445's & 2405's

I was wondering when a JBL guy would post about their polarity ! I guess they just wanted to be different than the rest of the world !
The 2445's & 2405's are black+ .....

What did you think about that diyaudio link I attached to my last post.... ?
A quick google search about JBL and series / parallel resistors turns up a ton of stuff.
Interested in what you think about buddy's solution...

M

 

[laatsch55]

JustMike where are ya bud??. He's our JBL guru..

 

[Michael F]

Correct ,it was mluciit the suggested moving the crossover points up... Anyhow,
I have the recently repaired a full WOPL 700B The one with "more output on the left channel" ...
After a rigorous run on the bench it came back upstairs to power Paradigm Signature S8's

Beautiful rich and silent.... This is using the passive crossovers within the S8's
I now have most of the tri amp stuff ready, just one more stock 400 to WOPL ....

As I'm tinkering with this new project I set the dsp for crossover points and do some basic configuration

WOPL 700B crossed @ 500hz 177L cabinet JBL 2035's
WOPL 400 crossed @ 7,000 JBL 2445's / 2390 horn + wave guide
Stock 400 crossed @ 7,000 JBL 2405's

This was my starting point... The DSP allows me to jump all over the place in real time with levels ,crossover points !

The DSP is driving the JBL's directly except for the addition of protection caps in series with the 2405's & the 2445's
No protect caps required for the LF 2035's
The selected protect caps won't have any effect on the set frequency response.....

The 2445's are 111 db efficient. The 2035's by contrast are 95 db efficient , now add the 2390 horn to the 2445's
and we effectively have more amplification (horn effect ?).....

So, the same amp will make more audible "idle" noise or hiss with no actual input signal when driving the 2 differently
efficient speakers ? Sound right ???

Any of the 3 amps I have ready experienced the same effect.... More idle noise with the more efficient driver.....
But, all silent with the Signature S8's and their passive crossover networks....

So, my thought was to use the above "l pad " idea to replicate what the passive crossover did without messing up
frequency response ?


Interesting read on the idea....


https://www.diyaudio.com/forums/multi-way/252713-cheap-free-improve-horn.html



I think I need a Sunday morning beer.....

M

It only stands to reason that hiss is a non issue with the S8 speakers (Paradigm?) and the 2035, the S8`s are 89db efficient and the JBL woofer doesnt climb that high.
You need to identify the source of the hiss. Is the hiss present with nothing plugged into the PL400s? If not, the source is further upstream. You can perform that test further up the chain until the offender is identified. Does the DSP operate +4 or -10 db and is it switchable? Are the inputs and output single ended or balanced?
As far as putting resistors in series with the compression drivers I would abstain. You already have a first order filter in those circuits by way protection caps ( make sure that the values are"invisible" in terms of your desired active dividing point).
There will be a certain amount of phase shift because of that but I would say its a necessary evil whose trade off is acceptable. Adding resistors will add more variables to the equation without dealing with the actual problem.

 

[Mohawk]

Paradigm S8'3 V3 ....
I believe the source of the "hiss " is the high efficiency of the compression driver & horn.... It is what it is, I'm just trying to reduce it givin what I'm working with...
The WOPL's are all RCA's
With or without any inputs ......
What do you think about the diyaudio post I referenced in post #10 ?
The protection caps have been selected using the Altec engineering chart ,attached below

M