Active Floor-stander

Hi guys,

I'm planning to document my journey in converting a diy 2 way floorstander into a 2.5way active crossover with dedicated 6 channel amplifier.

This project was ordered by my brother with a relatively good budget to start off with.
He auditioned speakers and amplifiers in the market costing up to RM20k and yet he's not happy with the sound so he's taking a gamble to ask me to make him a capable system without the high price tag.

more to come...

Well done Wikin, I suggest to your brother and to yourself to log on to www.avihifi.co.uk at their blog http://hddaudio.net/ to view a similar blog by a young contributor on the same subject.
Oh, by the way, I bought one of those ADM9ts and it sits in my house and I love the sound it makes. In fact I am now seriously pondering to purchase their newest active speaker - floor stander (ADM40)but at GBp3000 instead of the GBp1125 ADM9t

samazzah wrote:Well done Wikin, I suggest to your brother and to yourself to log on to www.avihifi.co.uk at their blog http://hddaudio.net/ to view a similar blog by a young contributor on the same subject.

Thanks for the blog. Will check it out.

For now I wanna make this m'sia boleh documentation la.



cheers

Starting off with the existing speaker:



This is a passive 2 way. With the white phase plug in place in the driver actually caused a lot of its bass power to be lost as the driver is leaky in air pressure. The thing I gained is mid range clarity.

So I'm going to add another 8" driver at the bottom of the cabinet. Driver at the bottom near the floor serves to add bass power much more as the floor reflection adds the qty of the bass.

Since the cabinet volume is now much smaller compared to the designed requirements, I'll have to perform full foam stuffing to gain back lost grounds.

For now this cabinet will be sent to the machine shop to get the hole machined out.

The tweeter is not too flushed onto the cabinet due to a thick gasket. So I will rectify this too by having the recess hole down by additional 3mm.

Drivers up to the task:

3/4" mylar tweet, 6.5" Kevlar, 8" Polyprop

They will cover the freq selected:

tweet: 3 - 20 khz
mid bass : 20 -3khz
bass : 20 -100hz

from my (limited) experience with conversion of a commercial loudspeaker to active drive, it's better to follow the slope of the original passive xover. replicate the slope in active and start from there. avoid lots of headache.

but since this is diy loudspeaker from the start i'm sure you have access to the driver spec sheet. as for me i have to re-measure the driver responses manually. lots of headache.

Hi, do you have any DIY Active or Passive Subwoofer Kit or any

info where we can buy the Kit ?

Wikin wrote:Drivers up to the task:

3/4" mylar tweet, 6.5" Kevlar, 8" Polyprop

They will cover the freq selected:

tweet: 3khz - 20 khz
mid bass : 20hz -3khz
bass : 20hz -100hz

The active x-over component will be using op-amps.

Design parameters were calculated using Elliott Sound's Linkwitz Riley calculator. It was very useful.



Mid to tweet x-over by right should use 24db for the clearest least muddy midrange; but due to the pcb space considerations, I decided to use 12db instead. I had a handful of 0.1uF capacitors (but upon measuring them they are around 95nF so that's a starting point for the calculator)

midrange


tweeter


woofer
Woofer x-over is using 100hz 24dB. This is important as I didn't want any mid range muddiness caused by the woofer's frequency overlap onto the mid range.

jat wrote:from my (limited) experience with conversion of a commercial loudspeaker to active drive, it's better to follow the slope of the original passive xover. replicate the slope in active and start from there. avoid lots of headache.

There is no hard and fast rule here bro, and I really don't take commercial design parameters seriously. All products are designed with cost in mind - and that applies for the choice of slope. The steeper slope will cost a lot of passive components and longer design times.

cheers

antaklugom wrote:Hi, do you have any DIY Active or Passive Subwoofer Kit or any

info where we can buy the Kit ?

Hi, I can only custom build the opamp circuitry for your active sub - sorry don't have a kit. Or you can check out ebay as there are several of china ones out there.

cheers

Wikin wrote:So I'm going to add another 8" driver on the bottom of the cabinet. Driver on the bottom near the floor serves to add bass power much more as the floor reflection adds the qty of the bass.

Since the cabinet volume is now much smaller compared to the designed requirements, I'll have to perform full foam stuffing to gain back lost grounds.

For now this cabinet will be sent to the machine shop to get the hole machined out.

Ahhh, the cabinets have just returned from the machine shop. Phew, what a mess.


And there are some chipped paint and scratches


Well at least the tweeter recess hole are done perfectly.



I made some new footers too.

Wikin wrote:
The active x-over component will be using op-amps.

Here comes the heart and soul of the system.

This pcb basically house the 2.5 way crossover with 12dB slopes between mid and tweet, and the 24dB of the woofer's slope.

One more twist I added into the crossover is a baffle step compensation. It makes the sound more correct to my ears because the narrow front baffle of the speaker will cause a 6dB rise after around 380hz. So it is only right to reduce that 6dB rise with this circuit. It will make the vocal thicker and more natural.

http://sound.westhost.com/bafflestep.htm

Just bear in mind to drive that baffle step compensation circuit with an opamp buffer.

A zoom into the left channel of the active crossover. I'm using dual opamp OPA2134. These opamps are really a class act in these applications. With the multiple opamp buffer requirements of the Linkwitz Riley circuitry, the opamp's output offset drift was very steady at 0.01 or so mV only.
That negates the use of interstage coupling caps to remove any residual DC components, hence simplifying the circuit and reducing component count.

Hi Wikin - good work. Always satisfying getting any DIY to work.

You may want to IC socket the opamps to allow some opamps rolling. I have earlier used OPA2134 (also TL072, NE5532 opamps) in my DIY 2-way crossover and have since moved to OPA627 (dual using brown dog).

I found an ebay source that sells them for RM20 + postage (dual opa627 with adaptor)! (Check the e-bay store - "hifiic" on the address below).

antaklugom - take a look at the url below for the PCB.

http://www.ebay.com.my/itm/2PCS-mono-adjustable-2-way-crossover-network-PCB-board-resistor-/320735064739?pt=LH_DefaultDomain_0&hash=item4aad4cb2a3#ht_2895wt_902

It is a 2-way linear-phase active crossover. The circuit works, but you would need to recalculated the values of the resistors/caps to meet your specific requirements.

Wikin wrote:A zoom into the left channel of the active crossover. I'm using dual opamp OPA2134. These opamps are really a class act in these applications. With the multiple opamp buffer requirements of the Linkwitz Riley circuitry, the opamp's output offset drift was very steady at 0.01 or so mV only.

For added clarity I've attached a block diagram of the crossover arrangement.
This crossover will be incorporated into an existing passive LDR preamplifier with source selector, hence making it a full fledge active preamp with remote control.



There are a total of 5 units of dual-op amp used per channel.

mthoi wrote:Hi Wikin - good work. Always satisfying getting any DIY to work.

You may want to IC socket the opamps to allow some opamps rolling. I have earlier used OPA2134 (also TL072, NE5532 opamps) in my DIY 2-way crossover and have since moved to OPA627 (dual using brown dog).

I found an ebay source that sells them for RM20 + postage (dual opa627 with adaptor)! (Check the e-bay store - "hifiic" on the address below).

Thanks MTHoi for the suggestion. Great minds think alike
That surely crossed my mind - if it hadn't been caused by a space constraint I would surely use the OPA627 in spades.

cheers

Wikin wrote:Drivers up to the task:

3/4" mylar tweet, 6.5" Kevlar, 8" Polyprop

They will cover the freq selected:

tweet: 3 - 20 khz
mid bass : 20 -3khz
bass : 20 -100hz

Hi Bro, mind to share what brand are those drivers?

Wikin wrote:
There is no hard and fast rule here bro, and I really don't take commercial design parameters seriously. All products are designed with cost in mind - and that applies for the choice of slope. The steeper slope will cost a lot of passive components and longer design times.

cheers

all drivers have a natural slope when reaching the edge of their frequency response. if you cross the frequency at the edge of their response you need to take the natural slope of the drivers into account.

4th order slope paired to the natural slope of the original drivers at 1st or 2nd order will lead to a steeper slope than you want. this will leave some big hole in the frequency response of the final speaker.

if you have access to the driver specs it's real easy to cross while the freq response is still flat thus avoiding the natural slope altogether. or cross with the slope taken into account.

time alignment of the drivers also will get changed after converted from active to passive. best if you could do close a mic measurement of the drivers to see how they would react in active. also it helps in final tuning of the driver level.

Wikin wrote:

There are a total of 5 units of dual-op amp used per channel.

The versatile "all-in-1" preamp looks like this from the gutt :


Front face was reworked. I added a black strip of plastic.


Align the position relative to the volume knob, selector LEDs





Voila, the preamp...




Rear panel:
Voltage input from the external PSU,
BMT means Bass, Mid, Tweet outputs


Inputs labelled 1,2,3,4 instead of some funky CD, DVD, TV, BR etc.


phew.... penatnya...

that looks like some serious work carried out. all inside a very narrow timeframe.

thumbs up for you bro.

Hi Wikin,
Very nice work and a round of applause from me.
regards
kp93300

atoz wrote:
Hi Bro, mind to share what brand are those drivers?

Hi bro,
The tweet is an SONY, mid bass is an AUDAX, bass from PIONEER. Land of the rising sun meets the west.

cheers.

jat wrote:
all drivers have a natural slope when reaching the edge of their frequency response. if you cross the frequency at the edge of their response you need to take the natural slope of the drivers into account.

4th order slope paired to the natural slope of the original drivers at 1st or 2nd order will lead to a steeper slope than you want. this will leave some big hole in the frequency response of the final speaker.

if you have access to the driver specs it's real easy to cross while the freq response is still flat thus avoiding the natural slope altogether. or cross with the slope taken into account.

time alignment of the drivers also will get changed after converted from active to passive. best if you could do close a mic measurement of the drivers to see how they would react in active. also it helps in final tuning of the driver level.

Correct on all counts bro.
In this project of mine I don't forsee any high risks of reaching the edge of the frequency response.
Well maybe just the tweeter is at slightly higher risk theoretically as it's only a 3/4" dome and not a 1".

cheers.

jat wrote:that looks like some serious work carried out. all inside a very narrow timeframe.

thumbs up for you bro.

Thanks for your support bro.

cheers.

kp93300 wrote:Hi Wikin,
Very nice work and a round of applause from me.
regards
kp93300

Thanks for your support KP. Lets have some teh tarik when you come over to KL.

cheers.

Continuation of the preamp build, that fella won't work without a dedicated power supply right?
So here's the matching psu in the making.

It started off with a blank plastic casing. I got it bored/ cut with the necessary holes for the IEC plug, a balanced socket (to output the 12v-0-12v ac) and the power switch.


I purchased a classic EI transformer which by luck, fits just nicely into the plastic casing. It buldges just a little that's unnoticeable with the lid closed.


Power switch was installed with an X-rated ( sounds obscene?) cap to absorb switch on clicks.


Earth was connected to the balanced output socket of pin 1 via a 10ohm resistor. Pin 2 and 3 is was connected to the 12vac(s).


Here's how it looks assembled and standing.



Powered it up unloaded - no fireworks and perfumes... AC voltage seems ok.

Ebony and Ivory.


PSU is linked via a standard AES/EBU interconnect.

Very interesting project, thanks for sharing.

Wikin wrote:
Correct on all counts bro.
In this project of mine I don't forsee any high risks of reaching the edge of the frequency response.
Well maybe just the tweeter is at slightly higher risk theoretically as it's only a 3/4" dome and not a 1".

cheers.

i learned that the hard way. bumpy road that i've traveled before. hopefully yours won't. i'll try and help as much as i can.

one more thing. i got into lots of trouble with hum in my previous endeavor. ground loop is a major concern. more so since i tried to integrate everything inside a single case.

Wikin wrote:
I made some new footers too.

So now it's back to working on the speakers. I started off with the new footers. They were made from furniture wood strips which are available at your local hardware store.

Each footer consist of a 2 piece part. They were glued and stacked on top of each other and then an additional weight was placed. It only took 1 hr to dry up.


The 2 piece part footers look like this.



Once dried, I drilled a pilot hole for an additional screw insert to add strength.


With the screw inserted, I drilled an additional pilot hole which will then be the positioning hole on the speaker's base plate.


With the footers locked in place, I further drilled onto the base plate.




Put on the spikes.




They stand

Onto the sponge foam stuffing.

I bought these from the local hardware store.

Cut, cut ,cut... plenty to do.


All dimensions are slightly oversized so that the sponge can stay in place.


That shiny material is actually the bitumen pad used for damping.


I hope I didn't over do it.

The internal speaker wires were built from a combination of CAT5 and twisted silver plated teflon solid core wire.

Raw material of wires were cut to length. The longest CAT5 wire was used for the bass driver. Mids and tweets driver will get the same length of wire. The long and circular black&white twisted wire is the silver plated teflon.


Strip baby, strip.



They were then twisted to form a high capacitance pseudo litz wire.

The cable for the mids and tweets driver get an additional treatment of silver plated teflon wires bundled together.


I then added some heat shrink.


Next comes the soldering process. The bare wire was soldered onto the mids and tweets driver as the drivers are quite mobile.



For the bass driver, the internal wire was first manually inserted into the speaker's internal body which was filled with sponge. It was a bit tricky but nevertheless I got the job done. Soldering commenced.


I layed down the speaker and then fastened the bass driver's mounting screw. While at it I also applied some black gloss paint onto the base plate. Hey it looks good as new.


Here's how it looks. It's still not done as I have yet to complete the tri-wiring speaker binding posts.

Wow - looks excellent. The build is certainly very interesting. Would you be kind enough to extend me an invite to listen to the finished product?

Thought I would share some of my findings on implementing my DIY crossover. I am currently in my third iteration of the crossover - tweeking op amps, power supply, coupling caps, etc. I have since :-

(i) removed the output buffer for the output channels since the OPA627 can drive the output lines quite well. This reduces one opamp on the circuit path. I read somewhere that the OPA2134 should be able to drive most outputs as well.

(ii) Place the volume control attenuator AFTER the crossover. There is a thread on the net about whether the volume attenuator should be "before" or "after" the crossover. I have tried both and preferred the set-up with the attenuator(s) placed after after the crossover. I am using a DACT-stereo to control the LF and Goldpoint-monos for the HF. (Making volume adjustments are more troublesome now - may implement a 4/6-gang volume control in the future).

(iii) Shunt regulated power supply at 18V for the OPA627 with individual coupling capacitors makes a whole difference.

Reg.

mthoi wrote:Wow - looks excellent. The build is certainly very interesting. Would you be kind enough to extend me an invite to listen to the finished product?

Thought I would share some of my findings on implementing my DIY crossover. I am currently in my third iteration of the crossover - tweeking op amps, power supply, coupling caps, etc. I have since :-

(i) removed the output buffer for the output channels since the OPA627 can drive the output lines quite well. This reduces one opamp on the circuit path. I read somewhere that the OPA2134 should be able to drive most outputs as well.

(ii) Place the volume control attenuator AFTER the crossover. There is a thread on the net about whether the volume attenuator should be "before" or "after" the crossover. I have tried both and preferred the set-up with the attenuator(s) placed after after the crossover. I am using a DACT-stereo to control the LF and Goldpoint-monos for the HF. (Making volume adjustments are more troublesome now - may implement a 4/6-gang volume control in the future).

(iii) Shunt regulated power supply at 18V for the OPA627 with individual coupling capacitors makes a whole difference.

Reg.

don't you need the output buffers to do some level adjustment to the different drivers? as not all the drivers have the same sensitivity, the output buffers allow some level equalization to be done so that the final speaker would have a flat frequency response.

Jat - The buffer stage is good for impedance matching - and the OPA627 seems to do drive the input of my DIY amps well without the buffer stage at the crossover.

Since I am using different DIY amps (solid state amp for the woofers, and tube amp for the mid/high), they are of different impedence, sensitivity, and the speakers have different drivers (105db/m woofers and 119 db/m horns), it make sense have individual attenuators for the amps to control the volume.

Unfortunately, getting a "flat" response curve is a "problem" for each setting of volume since there are 3 volume controls. But it reaches a point where I just settle on a 2db difference between the HF/LF volume setting.

With my old ears, I like to have the HF loud(er) - maybe a bit piercing for normal hearing, but that is the joy of hi-fi.

Wikin wrote:
Here's how it looks. It's still not done as I have yet to complete the tri-wiring speaker binding posts.

While the speakers are lying down on the floor I took the opportunity to improve the paintwork and appearence.

Some black paint was dripped into the tweeter and woofer's screw holes to make them look more flush. Since there is anticipation that I would not be opening them up for quite some time so heck.


After 1 day the paint dried up. It looks pleasing actually.


Same goes for the woofer's screw. Once they are dried they look almost seamless from a distance.


After 1 day of dry time they are standing up again. Now comes the work on the rear binding posts.


The original binding posts are bi-wire type and since this speaker is now tri-wire, some thoughts arise. Should I just drill another 2 holes and add some posts onto this fella? Where's the best place to position the 2 binders so that my fingers can access them properly?



Nah, I'll keep this speaker post for another project.
So I dished out some plastic plate I had in my junk box and behold, there were a big bunch of new posts that I bought the last time.


Since we are almost close to powering up this speaker to hear her sing, the 'quick n dirty job' is the rule of the day. Furthermore this is on the rear backside - nobody's gonna notice some crude work anyway. Off it goes to the jig saw.


I measured and drew up a piece of paper mask that will allow the 6 speaker binding posts to be spread nicely on the plastic plate so that tightening them up will be ergonomically friendly. Don't you just hate it when some speaker posts are spread too closely apart it hurts your finger sometimes?




The pilot holes made the drilling job easier as the drill bit will not slip. Starting with the finest bit, then slowly working up to a larger diameter, the plate was completed.


Wallah...

mthoi wrote:Wow - looks excellent. The build is certainly very interesting. Would you be kind enough to extend me an invite to listen to the finished product?

I will surely do thanks for asking.

mthoi wrote:
(i) removed the output buffer for the output channels since the OPA627 can drive the output lines quite well. This reduces one opamp on the circuit path. I read somewhere that the OPA2134 should be able to drive most outputs as well.

Noted on this and it's great news. Just in case if the 12db midrange crossover does not sound too satisfactory I will convert the output buffer into another 12db slope hence effectively making them 24db/octave.

mthoi wrote:
(ii) Place the volume control attenuator AFTER the crossover. There is a thread on the net about whether the volume attenuator should be "before" or "after" the crossover. I have tried both and preferred the set-up with the attenuator(s) placed after after the crossover.

Ah yes, I've implemented this on my Open Baffle active system which serves as my main system too. Speaking of which, I do have a spare 6 gang Alps pot. Lets see if this project turns out well or not i.e. if the clarity is satisfactory with the LDR.

mthoi wrote:
(iii) Shunt regulated power supply at 18V for the OPA627 with individual coupling capacitors makes a whole difference.

Which model of the shunt regulator did you use?

cheers.

I think the spelling for the brand "Pionneer" vs "Pioneer" don't seem to be right!

atoz wrote:

I think the spelling for the brand "Pionneer" vs "Pioneer" don't seem to be right!

Sharp eyes you have there friend.
Initially I wanted to use these as bass helper which my brother was okay with it.

After giving some deeper thoughts, I didn't like the idea of using a 4 ohm driver as a bass unit as I like to give the amplifier an easy & light load/working conditions.
So off I went scouting for some larger 8ohm drivers and this fella came about. It's a clone which can do some decent air pumping when tested.

cheers

I used to collect speaker and drivers since a very long time ago, it is more like a hobby to me rather than an real audiophile so i have come across all sorts of genuine and imitation or clone drivers, that's why i am a bit sensitive to the authenticity of a product.



But the bottom line is that the quality of the product itself that counts, not the brand.



Happy DIYing!

Wikin wrote:

The good thing about having a lot of real estate between the binding posts is that passive components can be soldered directly here hence this board serves 2 functions.

But wait, what passive components are we talking about - isn't this an active speaker project?

Take a guess.


Guess some more.... from left to right - bass, mid, tweet.


The binders are now soldered to the speaker's tail end wires.


Now comes the screwing part. This is a bit tricky as the wires are stiff, so I had to gently move all 3 sets of wires tucking them back into the speaker box while getting strong resistence from the thick internal foam.


Easy does it...


Whew, done it. Now onto the other side of the speaker...

Hi Wikin,
What is the effect of the Zobel network in sound in your experience?
There is no zobel in my speaker / amp and curious to try.
How do you calculate the values of the C and R ?
thanks for showing your work and learn something today !
regards
kp93300

So now it's onto building the amplifiers.

I've selected LM3875 since I have a bunch of them on hand from a previous project and they do sound good when properly implemented.


Initially I built the NIGC circuit ala Gainclone style - alas I couldn't stablize them as they were just noisy and the DC offset wasn't good at around 80mV.

After that I surfed around for another circuit and ended up with a standard IGC using 220Kohm feedback resistor.


The single bridge rectifier was mounted directly below the main capacitor where its +ve and -ve legs were bent 180degrees and soldered onto the capacitors.


This is for the bass module. The capacitance was increased to +/-3000uf just in case. The other modules for mids and tweets were standard +/-1000uf.


Lots of work and tedious job when building up 6 of them actually.

Here's 4 channel being mounted onto a temporary heat sink. Capacitor bypass is a must for gainclones.




Fast forwarding, the 6 channels were up. From bottom row to top - the mid amp, bass amp, tweet amp.


The IGC worked well upon 1st firing up. To my surprise the DC offset was excellent at 0mV !

kp93300 wrote:Hi Wikin,
What is the effect of the Zobel network in sound in your experience?
There is no zobel in my speaker / amp and curious to try.
How do you calculate the values of the C and R ?
thanks for showing your work and learn something today !
regards
kp93300

Hi KP,
I basically use this calculator:
http://diyaudioprojects.com/Technical/Speaker-Zobel/

Using an LCR meter the values of the required parameters can be measured. Just measure the coil inductance and the total coil resistance of the speaker driver and then plug in the values.

As for the sound, do bear in mind that if your existing crossovers have been designed without these zobel in the beginning, chances are if you add them now, it will muck up the sound for the worst.

OTOH if the zobel was there in the 1st place, the effect would be a much more stable sound when playing at very high volumes.

cheers

Hi Wikin
Thanks for the pointer
Alas, I do not have an inductance meter .
kp93300

Finally it's judgement day and time to power up the whole she-bang together to have a listen.... AND.... do you know that feeling you just wanna SHOUT M.C.B!!!

There were several problems:
1) The dreaded GROUND LOOP is buzzing like mad.

2) One of the bass module is not working

3) When I crank up the volume the mids and woofer started to crackle.

Debugging session.

Started with item #2.
So I unplugged the bass module, changed the LM3875 chip and it's back working. I suspect a bad solder joint because the LM3875 chip was in fact in working condition after I re-tested the suspected chip.

Item #3.
The crackling sound is basically distortion and only appears during bass heavy passages ESPECIALLY Hotel California's opening track.
So I was thinking it's either the active XO module distorting or the Gainclones are clipping.

Earlier on the active XO modules were tested bare board into a 2-way system before the bass opamps were developed. The 2-way sounded pleasant without any issues and at that time a different amplifier was used.

Therefore most likely the issue should be from the Gainclones. Nevertheless just in case Murphy says otherwise, I took some extra precaution to add extra 3300uF 35V capacitor on the active XO's DC supply across the pins of + and - . Also since the active XO's PSU is unregulated +/-16VDC, I added a CRC filtration. There's plenty of small 0.01uF capacitor bypass directly below the opamp's supply pins.

Further cleaning up of the active XO module's PSU is adding capacitor bypass in delta arrangement across L-N-E (grey caps) and also across the 12VAC-0-12VAC (blue caps)



Earlier on, the earth pins was connected to the 0V of the transformer. I removed it and thought to perceive a lesser buzzing noise.

Later after reading and researching more regarding ground-loop problems in audio systems, I decided to re-install the earth connection. This is because, according to the Jensen literature, removing earth connections are a hazard to safety and it's not really solving the real problem, just a patch job.

http://www.jensen-transformers.com/an/generic%20seminar.pdf


The result : STILL BUZZING LIKE MAD.

Doing some reverse troubleshooting following the Jensen literature, I started at the amp's interface.

When I removed all interconnects from the 6ch amp, it's very quiet without buzz. Well at least the amps are built to a respectable standard I thought.

Installing the RCA portion by portion, when only a pair of bass, midrange or tweeter's RCA was installed at 1 time, there's still quietness.

Next I installed the mid's RCA together with the bass's RCA. Light buzzing started to appear.

Finally when I installed the tweet's RCA, wow it started to BUZZ like mad.

Therefore the conclusion is that the grounding path needs further work and most likely it's the amplifier's ground path that's the culprit from experience.

I'm suspecting less of the active XO module because from the way the Active XO module was built, the grounding is a solid plane on the output and input RCA section tied together.

Targeting the 6 channel amp's ground path, it's now trial and error since it's quite difficult to grasp the concept of improving the ground path - whatever that means.

So I decided to try and improve the power supply and grounding cable by increasing its gauge.

The 6 channel power supply cable that supplies the 25V-0-25V was rebuilt using standard 18AWG power cord. Out goes the single core cat5.



Strip, strip, strip,


Twist, twist, twist, solder, solder, solder.
My finger tips started to sore.


Finally installed into the amp modules.


Power on and testing commence.
Hey much lesser buzzing now. But still, there's buzzing sound that's not as quiet as it should be. From the seating position about 6 feet away, the buzzing is still faintly audible. That's not good enough.

On a side note, the midrange crackle during bass heavy passages in Hotel California was completely eliminated! Therefore the conclusion was that this issue must be a current starvation problem that was solved with a thicker gauge supply cable. How silly of me to use a single cat5 out of convenience DUH!

The bass sound has improved significantly but still some small level of crackling can be perceived. Excited with the progress, I changed the rectifier of the bass amp to this huge bridge capable of 25A.


PROBLEM SETTLED!!


Ok so one last final hurdle is back to solving the ground loop problem. I believe this can be done as my main system which is also a 3-way active system is humm free.

Hi Wikin - well done on all the positive progress in the build.

In my last active crossover build, I have similar grounding issue that causes hum with the interconnects. I realised that the signal ground is hanging; my humming problem was solved when connected to the power supply ground/chassis ground (at a single point).

Don't forget to invite for a listening session when you complete the project. Reg