Definitive Technology (Def Tech) BP7006/BP7004/BP7002 - BP8954MOD

[datrumole]

Hey man! Loved the write-up, went ahead and tried this myself with a BP7004. To my surprise I didn't break everything lol. However I am getting a ton of ground noise (i think) when the amp is on with nothing connected. If i connect to a source, the hum gets lower but is still quite noticeable. Any tips?

Thanks!

hey, thanks!

yeah, i get a bit of hum too. someone smarter than me might be able to comment. it's possible that some caps in front of the amp might help some of the noise coming from the transformer? or maybe upgrading the caps on the board itself? i'm far enough away that i dont really hear it on standby. i'm not sure if it's the amp module or the transformer making the noise. i may have a new oscilliscope xmas time that might allow me to dig further to find out

 

[windowman]

So I've had these speakers since 2007 and started to have noise problems. Saw this thread and decided to try these amps, using the 24v secondaries on the transformers. First I hooked them up to a cheap, small pair of speakers to evaluate. The hum and resudual noise is too high for me. They work, but are not ready for prime time. I added 4,700uF of power supply filtering to each rail...no difference. Did a few quick measurements with the added caps. There's 12mV of AC noise/ripple on the + rail. There's 4.6mV of AC on the speaker out (with no load connected) with the pre out off but physically connected, and 6.0mV with the input shorted. And 2.5mV AC with an open circuit input. Not sure what else to try to get the noise down. Any suggestions?

 

[datrumole]

So I've had these speakers since 2007 and started to have noise problems. Saw this thread and decided to try these amps, using the 24v secondaries on the transformers. First I hooked them up to a cheap, small pair of speakers to evaluate. The hum and resudual noise is too high for me. They work, but are not ready for prime time. I added 4,700uF of power supply filtering to each rail...no difference. Did a few quick measurements with the added caps. There's 12mV of AC noise/ripple on the + rail. There's 4.6mV of AC on the speaker out (with no load connected) with the pre out off but physically connected, and 6.0mV with the input shorted. And 2.5mV AC with an open circuit input. Not sure what else to try to get the noise down. Any suggestions?

yeah, the amp certainly suffers from a slight hiss. i'm sure someone more competent on amp building would know how to adjust/fix

common fixes or improvments i see mentioned to improve not-so-great chinese boards: filtering on the AC from the trans, filtering post dc-rectifier (upgrade the board caps), output coils (no idea what they actually do)

might be worth a thread on the diyaudio board as they generally tend to have more EE type folks frequent there. could try here too, shoot me a link to the thread if you ever create one

i found that the hum was so minimal once connected to the sub on the speaker, that it didnt much matter

 

[windowman]

yeah, the amp certainly suffers from a slight hiss. i'm sure someone more competent on amp building would know how to adjust/fix

common fixes or improvments i see mentioned to improve not-so-great chinese boards: filtering on the AC from the trans, filtering post dc-rectifier (upgrade the board caps), output coils (no idea what they actually do)

might be worth a thread on the diyaudio board as they generally tend to have more EE type folks frequent there. could try here too, shoot me a link to the thread if you ever create one

i found that the hum was so minimal once connected to the sub on the speaker, that it didnt much matter

YES, that diy thread is right here: https://www.diyaudio.com/forums/class-d/333317-tda8954th-btl-amplifier-2.html
The key post is #12, which points out 2 HUGE flaws on the original board: wrong parts values. The small caps in the snubber circuits, and the power supply decoupling resistors. On my boards the caps appear to be ok (if they're bad the partnering resistor blows open. Mine didn't). BUT my resistor values were wrong. They're supposed to be 10 ohms, but were only 1. I can see how a higher value would "burn off" some noise, so I'll make the changes as soon as my new soldering iron comes in. (Murphy strikes again!). These resistors are tiny surface mounts, and I need a fine tip or else it'll be a disaster. Can't wait to get this done and report on how it works. Those parts BTW are under the heat sink, so a careful hand is a must.
You'll notice that this person ("FauxFrench") does a WHOLE lot more to this board that I wont do, like turn it into a 2-channel unit! But the info is there for those so inclined. Stay tuned...

 

[datrumole]

YES, that diy thread is right here: https://www.diyaudio.com/forums/class-d/333317-tda8954th-btl-amplifier-2.html
The key post is #12, which points out 2 HUGE flaws on the original board: wrong parts values. The small caps in the snubber circuits, and the power supply decoupling resistors. On my boards the caps appear to be ok (if they're bad the partnering resistor blows open. Mine didn't). BUT my resistor values were wrong. They're supposed to be 10 ohms, but were only 1. I can see how a higher value would "burn off" some noise, so I'll make the changes as soon as my new soldering iron comes in. (Murphy strikes again!). These resistors are tiny surface mounts, and I need a fine tip or else it'll be a disaster. Can't wait to get this done and report on how it works. Those parts BTW are under the heat sink, so a careful hand is a must.
You'll notice that this person ("FauxFrench") does a WHOLE lot more to this board that I wont do, like turn it into a 2-channel unit! But the info is there for those so inclined. Stay tuned...

yeah, let me know how the mods go and i'll get them posted into the main post! great find!

i sold my speakers, so i'm not able to do any further mods

 

[Stu Pidasso]

I stumbled on this this thread just now and thought I could add a little here for future DIY'ers. Simply replacing the entire board on this loudspeaker will not provide the best results in my opinion, as there is equalization and crossover that are specific to this loudspeakers operation. The ones I am repairing for a friend have traces beneath the supply caps vaporized and the mosfets are shorted on the amplifier board. Upon removing the heat sink from the mosfets I see that they have had their numbers obfuscated. Jerks. I do happen to have a schematic which I'll post here - the schematic covers everything but the class D amplifier section - where the problem is occurring. The schematic I got from the SoundUnited service guys has a bunch of handwriting on it indicating an International Recitifier 331(5 or J) which is NOT the correct mosfet. What I have decided to do, since SoundUnited doesn't even have replacement parts for this any longer is take the post eq/x-ovr signal from the pre-board and connect that to a class D board that runs on +/- 66 volts which is what one of the sets of windings delivers. Additionally, one of the diodes in the 66v full wave rectifier is open, and if I did not replace the rectifier an obnoxious hum would occur - perhaps this is what the previous guy was going through on his speakers.
Finally, I would suggest you find a class D board that is advertised to run with not only a +/- 70v max supply but one that puts at least 200w into 8 ohms. So many of the boards I've looked at claim preposterous outputs but upon close inspection you'll see that the claimed wattages are only at .5 ohms at 10% THD for 1ms or some such nonsense. I've settled on a 600w module from Ebay that runs on +/-70v max. I won't link it here because I don't want anyone to buy this thinking I am co-signing for this board. The board I'm going to try uses the IRS2092S class D driver and runs well on high voltage at high speed. Be sure the one you determine to purchase is not using an open choke. The one you want has a enclosed ferrite output inductor. The original board includes a mute input to keep the amplifier board from making a popping noise when the preamp takes longer to ramp up than the class D board. This is an easy mod, you can either use that 15v mute line to switch a pair of mosfets that you can install in the supply rails to the amplifier, or you could use it to turn on a relay which cuts the speaker outputs from the amplifier. Your call.

 

[datrumole]

I stumbled on this this thread just now and thought I could add a little here for future DIY'ers. Simply replacing the entire board on this loudspeaker will not provide the best results in my opinion, as there is equalization and crossover that are specific to this loudspeakers operation.

that is correct, by replacing the board, you are foregoing the ability to use the 'speaker level' inputs

i dont have a working board to compare if the board does anything to the LFE signal, or the EQ is just there for the speaker level input. i think we can confidently say it has some filtering/eq for the speaker level input as it needs to hit it with a LP120hz (or whatever they've set it to, likely somewhere in the 60-200hz range) and possibly a subsonic filter as well for sub20hz. but unknown if the same level of filtering would happen on the LFE input as you'd end up doubling up the LP filters causing a phase shift and an even steeper roll off, so my guess is the LFE is likely untouched

so after this mod, since you have to use LFE, i would hypothesize you aren't missing out on any filtering or EQ, and if you have audyssey with sub EQ, irrelevant as you'd gain it back anyway

again, unless someone has a working one and can do a measurement of the individual driver response, and compare it to a bypassed board, would be the only way to prove it.

Finally, I would suggest you find a class D board that is advertised to run with not only a +/- 70v max supply but one that puts at least 200w into 8 ohms. So many of the boards I've looked at claim preposterous outputs but upon close inspection you'll see that the claimed wattages are only at .5 ohms at 10% THD for 1ms or some such nonsense. I've settled on a 600w module from Ebay that runs on +/-70v max. I won't link it here because I don't want anyone to buy this thinking I am co-signing for this board. The board I'm going to try uses the IRS2092S class D driver and runs well on high voltage at high speed.

everyone is certainly welcome to get whatever module they want, however the larger boards, require larger power supply stages, which complicates the entire build where you are competing for limited space if you are trying to keep it mounted to the plate, but go for it! the IRS2092S boards are excellent amps (and likely overkill for just sub duty)

yes, these boards often times have inflated spec's (you think the amp board you are pulling out really has 300w ) , but depending on which module you have 8/10/12" driver, you'd be surprised how little wattage you'd actually need. with the little class D amp i originally suggested, i have to set to -3Db on the LFE output as i'm able to bottom out the driver.

i cant say we know the spec's for the subs in these, so to give a requirement to say you need 200w into them might end up with some blown drivers, but again, would need to test their power handling before i can confidently say you can't

thanks for chiming in!

 

[Stu Pidasso]

Actually, I do know what's going on in the pre-amp of the stock board. There is a limiter which begins to kick in at 28.28vrms and goes into full clamp mode at 77v peak - or if the input voltage is higher than 2.9956vrms when using the RCA input. 28.28v rms into 4 ohms is 200 watts. There is a high pass filter at 24db per octave at 28.46hz, an EQ bump at 35hz Q-3, 8.465db, phase adjustment is also present. From 50hz to 150hz we have a shelf with a q of .784, D=1.2755, 4.771db. We also have a low pass filter at 150hz which is strange as it's 18db per octave.

Without these points taken into consideration, you're not dealing with the same speaker by simply adding a class D board amplifier.

 

[Wseaton]

Good thread. Great tips.

Still, I wish DT had stuck to passive designs. I used to work at a high end audio dealer and vividly recall DTs passive BP 20 / 10 and their miraculous BP 2 rears. IMO, bi poles for rears, provided you had an appropriate room are still the top of the food chain for rear channel seamless sound.

I never liked DTs move to powered speakers and hated the BP2000s. Then again what do I know.

 

[datrumole]

Actually, I do know what's going on in the pre-amp of the stock board. There is a limiter which begins to kick in at 28.28vrms and goes into full clamp mode at 77v peak - or if the input voltage is higher than 2.9956vrms when using the RCA input. 28.28v rms into 4 ohms is 200 watts. There is a high pass filter at 24db per octave at 28.46hz, an EQ bump at 35hz Q-3, 8.465db, phase adjustment is also present. From 50hz to 150hz we have a shelf with a q of .784, D=1.2755, 4.771db. We also have a low pass filter at 150hz which is strange as it's 18db per octave.

Without these points taken into consideration, you're not dealing with the same speaker by simply adding a class D board amplifier.

so when using an RCA LFE input, they are double filtering? sounds like a bad design then and one should NEVER use the RCA LFE input as the LFE will already have those filters built in, and likely be unconfigurable. changing the phase and rolloff. ouch! bad def tech

 

[Stu Pidasso]

so when using an RCA LFE input, they are double filtering? sounds like a bad design then and one should NEVER use the RCA LFE input as the LFE will already have those filters built in, and likely be unconfigurable. changing the phase and rolloff. ouch! bad def tech

Review the schematic I posted earlier in this thread.

 

[datrumole]

Review the schematic I posted earlier in this thread.

yeah, all good, i'll take your word for it

so, to finalize your stance, if someone uses RCA input, which by right has an already existing 120hz LP and likely a 20hz HP on their LFE output from their receiver, people are getting double filtered and phase issues done to their woofer, and they've been selling this for 20+y and no one but you has noticed?

 

[Stu Pidasso]

I think you may be missing the point. Look at the schematic and you'll understand what's going on. The LF filter is simply tuning the woofer in it's enclosure for the designers desired response when mated with the upper speaker drivers. These woofers are NOT subwoofers. They are bass drivers and do not begin to cover proper LFE spectrum response.

 

[Hecdtec]

I was going to attempt to do the amplifier replacement as I purchased the amps but there still are a few issues with this modification. 1. What to do about the humming issue. 2. Need to add a pot to control the volume. 3. There is no protection for the woofer if the amp has issues just like the original ones don't. Instead I'm deciding to go with an external amp to resolve all these issues. I picked a Crown XLi 1500 with 330w x 2 @ 8 ohms/450w @ 4ohms. Input will be from the sub out from the home theater receiver and controlling power to the Crown using a smart surge suppressor that uses a master outlet plug where home theater receiver will be plugged into which controls the other outlets which the Crown will be plugged into. I also will be drilling holes in the speakers amplifier plate to install a set of binding posts and disconnecting the woofers wiring from the built-in amplifier to the new binding post internally of the speaker which will be wired to the Crown externally of the speaker. Obviously the built-in amps will be unplugged and cable will be cut off when it's all working. I will let you know how this works out as I just ordered the Crown from AudioSavings.com for $249 open box along with the insulated binding posts from Amazon. I will have to play with the sub out frequency to get the crossover point from midwoofers just right and I'll report back with other part numbers, place of purchase and pics Wish me luck.

 

[Stu Pidasso]

Regarding the humming issues, you should check the power supply in the existing speaker, ensure the full wave rectifiers are working properly. Also a simple swap out of the caps may be in order. If you use an external amplifier, you will want to ensure the same equalization the manufacturer designed into the pre-driver circuit in the OEM device. You can see those settings in the schematic I posted in this thread.

 

[datrumole]

I was going to attempt to do the amplifier replacement as I purchased the amps but there still are a few issues with this modification. 1. What to do about the humming issue. 2. Need to add a pot to control the volume. 3. There is no protection for the woofer if the amp has issues just like the original ones don't. Instead I'm deciding to go with an external amp to resolve all these issues. I picked a Crown XLi 1500 with 330w x 2 @ 8 ohms/450w @ 4ohms. Input will be from the sub out from the home theater receiver and controlling power to the Crown using a smart surge suppressor that uses a master outlet plug where home theater receiver will be plugged into which controls the other outlets which the Crown will be plugged into. I also will be drilling holes in the speakers amplifier plate to install a set of binding posts and disconnecting the woofers wiring from the built-in amplifier to the new binding post internally of the speaker which will be wired to the Crown externally of the speaker. Obviously the built-in amps will be unplugged and cable will be cut off when it's all working. I will let you know how this works out as I just ordered the Crown from AudioSavings.com for $249 open box along with the insulated binding posts from Amazon. I will have to play with the sub out frequency to get the crossover point from midwoofers just right and I'll report back with other part numbers, place of purchase and pics Wish me luck.

no luck needed, the built-in amps aren't magic, they are just amplifiers

looks like you are headed down a perfectly fine path, external amps work just fine, just more costly than the mod

enjoy the revival of your def techs!

 

[Stu Pidasso]

no luck needed, the built-in amps aren't magic, they are just amplifiers

This is what I keep saying is not accurate. Unless you take the post electronic crossover, limiter, eq, and phase adjustments applied by the OEM system you will NOT be anywhere near close to the original response. Not trying to pick a fight with you datrumole but your statement is not accurate given the context.

 

[Hecdtec]

Well with a few issues I had to resolve I finally got it to work.

1. I had to decide on the only binding posts from Amazon tha I can visually see that was insulated to install through a metal plate.
2. I removed the amplifier and unbolted the toroidal transform and 3. it's bracket.
4. Measured and drilled the holes so that they would line up to a dual banana plug.
5. & 6. Measured, cut, stripped and terminated a 12 gauge oxygen free wire that went to the binging posts to the woofer.
7. Repaired one of the post with a thicker nut so it used more of the threads.
8. Wired up the speaker, setup the amp for parallel input from the single subwoofer output.

I could not get the subwoofer to work once I added the subwoofer to the speaker configuration on the Receiver. Got the humming from the woofer when moving the rca cable from subwoofer 1 to 2 outputs but still no sound. Finally researched (Googled) the issue and found I either had to set the fronts to small speakers to get the signal to go to subwoofer output or as I found at a reply to same post that setting the double subwoofer in the crossover section of speaker config meant subwoofer frequency will be fed to both front Large speakers and the subwoofer speaker output which got the subwoofer and connected amp to finally come alive.

Datrumole I want to thank you for inspiring me to work on a solution by your successful attempt with a replacement amplifier. I chose the path I went to ease the modifications to the board electronics, addition of potentiometer and mounting it all. I also wanted to protect the woofer from amplifier malfunction that is built into this Crown professional amp. The cost to have Def Tech repair an already aging single amplifier that could break down anytime for $200 - $300 for one amp instead gave me 2 new amplifiers I could use for the speaker pair from the Crown amp. Getting a 6 years warranty on Crown amp is another benefit.

Stu Pidasso if you can figure out at what frequency the original amps crossed over would be much appreciated. I wrote Definitive support and they said that they no longer had the documentation on the BP 7000 series and to try using 80-100hz. I can't believe how a company that built a product doesn't know where or doesn't want to disseminate that information.

If there's any specifics that anyone needs reach out to me and I will get back to you on the specifics of products I used or on the install. Thank you gentlemen and good luck with your replacement configurations.

 

[datrumole]

This is what I keep saying is not accurate. Unless you take the post electronic crossover, limiter, eq, and phase adjustments applied by the OEM system you will NOT be anywhere near close to the original response. Not trying to pick a fight with you datrumole but your statement is not accurate given the context.

no worries, i respect your opinion on the matter. my opinion is that all of what you are talking about is likely bypassed when using the RCA-in. anyway, i dont have the speakers anymore to provide some tests and measurements, nor did i ever have a working amp. so i'd have to rely on someone willing to give it a shot. until then, you hold your opinion, i hold mine. its ok. you are welcome to provide some measurements and proof of yours and i'd be happy to change mine

Well with a few issues I had to resolve I finally got it to work.

1. I had to decide on the only binding posts from Amazon tha I can visually see that was insulated to install through a metal plate.
2. I removed the amplifier and unbolted the toroidal transform and 3. it's bracket.
4. Measured and drilled the holes so that they would line up to a dual banana plug.
5. & 6. Measured, cut, stripped and terminated a 12 gauge oxygen free wire that went to the binging posts to the woofer.
7. Repaired one of the post with a thicker nut so it used more of the threads.
8. Wired up the speaker, setup the amp for parallel input from the single subwoofer output.

I could not get the subwoofer to work once I added the subwoofer to the speaker configuration on the Receiver. Got the humming from the woofer when moving the rca cable from subwoofer 1 to 2 outputs but still no sound. Finally researched (Googled) the issue and found I either had to set the fronts to small speakers to get the signal to go to subwoofer output or as I found at a reply to same post that setting the double subwoofer in the crossover section of speaker config meant subwoofer frequency will be fed to both front Large speakers and the subwoofer speaker output which got the subwoofer and connected amp to finally come alive.

Datrumole I want to thank you for inspiring me to work on a solution by your successful attempt with a replacement amplifier. I chose the path I went to ease the modifications to the board electronics, addition of potentiometer and mounting it all. I also wanted to protect the woofer from amplifier malfunction that is built into this Crown professional amp. The cost to have Def Tech repair an already aging single amplifier that could break down anytime for $200 - $300 for one amp instead gave me 2 new amplifiers I could use for the speaker pair from the Crown amp. Getting a 6 years warranty on Crown amp is another benefit.

Stu Pidasso if you can figure out at what frequency the original amps crossed over would be much appreciated. I wrote Definitive support and they said that they no longer had the documentation on the BP 7000 series and to try using 80-100hz. I can't believe how a company that built a product doesn't know where or doesn't want to disseminate that information.

If there's any specifics that anyone needs reach out to me and I will get back to you on the specifics of products I used or on the install. Thank you gentlemen and good luck with your replacement configurations.

congrats bud, glad i could help

as for the crossover freq, if you have a mic and rew, run a sweep with just the upper array connection, it should be fairly obvious when it rolls off. or even use your phone with an SPL app. again only the upper array, play 200hz, note the reading, play 190hz, note...etc all the way down to like 80hz. while it may not be accurate in terms of actual SPL, it will be relative to one another, so it will work just fine for this use case. you'll then be able to see where it rolls off. when you've reached a -3 to -5db, should give you a good enough indicator what the crossover point should be

once you set it, you can then flip the phase back and forth and see which one gives you the better reading on the mic as well

 

[Stu Pidasso]

Hey I posted the schematic and on each page at the top you may note what it is that particular circuit section is doing. For instance, you may see 35hz - Q-3.00 +8.34db = this would equal a boost at 35hz of 8.34db with a Q of 3. Each page has a little bit of information, I'm not going to list it out for you, check out the schematic and each page will tell you a little bit more in plain english at the top of the page.