Hi,
I'm building an active speaker and I'm almost certain what components I'm going to use. I would just like to double check here if anyone sees any problems with my setup.
Im powering a 30W RMS full range and a 15W RMS tweeter, I will probably not push this unit very hard, since I will be at a very close distance from it.
Bi-Amp/X-O: Audio Bi-Amplifier, 40W w/ Linkwitz-Riley Crossover - Unbalanced Input [XAMP-M2]
PSU: DC Power Supply, 50W, Fixed or Variable Voltage [XAPS-50]
Transformer: Transformer 48VA 2x12V 2A (in Swedish, but you get the idea)
Any input is greatly appreciated.
I'm building an active speaker and I'm almost certain what components I'm going to use. I would just like to double check here if anyone sees any problems with my setup.
Im powering a 30W RMS full range and a 15W RMS tweeter, I will probably not push this unit very hard, since I will be at a very close distance from it.
Bi-Amp/X-O: Audio Bi-Amplifier, 40W w/ Linkwitz-Riley Crossover - Unbalanced Input [XAMP-M2]
PSU: DC Power Supply, 50W, Fixed or Variable Voltage [XAPS-50]
Transformer: Transformer 48VA 2x12V 2A (in Swedish, but you get the idea)
Any input is greatly appreciated.
48VA is a bit low.
Normally I recommend anywhere from 1times max total output power to 2times max total output power.
that would give 40+15 = 45 so a 45 to 90VA transformer should work adequately for one channel.
But a 48VA transformer will have a high transformer regulation value. This high regulation will raise the no load output voltage supplied to the amplifiers a lot maybe 15 to 20% compared to the maximum power supply level.
All your components must be rated to tolerate the highest output voltage from your transformer when mains voltage is at it's highest and the loading is at it's lowest.
Just looked at your link. That is the same manufacturer as another Member linked us to.
I thought the single ended supply and the bridged topology was terrible for an active speaker implementation. And you plan to use a regulated supply !
You clearly see price as over-riding other concerns.
Normally I recommend anywhere from 1times max total output power to 2times max total output power.
that would give 40+15 = 45 so a 45 to 90VA transformer should work adequately for one channel.
But a 48VA transformer will have a high transformer regulation value. This high regulation will raise the no load output voltage supplied to the amplifiers a lot maybe 15 to 20% compared to the maximum power supply level.
All your components must be rated to tolerate the highest output voltage from your transformer when mains voltage is at it's highest and the loading is at it's lowest.
Just looked at your link. That is the same manufacturer as another Member linked us to.
I thought the single ended supply and the bridged topology was terrible for an active speaker implementation. And you plan to use a regulated supply !
You clearly see price as over-riding other concerns.
The problem that I see is that you have a rather low wattage system. It turns out that most low VA (less than 150VA) transformers have very poor voltage regulation because manufacturers (in my experience) very liberally assign it or perhaps that is just how those low VA transformers happen to function. For example, even though your transformer might state "2x12 48VA" what you get is likely to be something like:
no load voltage: 16V
2A voltage 12V
The difference between the no load voltage and the loaded (at rated current) voltage is related to the voltage regulation. There does not seem to be a standard for exactly how the no load and loaded voltages are assigned by manufacturers, so unless the MFG provides test results you will have to do it yourself.
It may be the case that the no load voltage exceeds what the amp module can safely accept. If it is the no load voltage that is the rated voltage, then the amount of voltage sag under the rated current might be unacceptable. Either case is bad. The way to prevent this problem is to use a transformer with a higher VA rating. High VA transformers have better voltage regulation AND, since you will not use up the full (average) current rating, the amount of voltage difference between no load and 100% load will be again less than for a lower VA rated transformer.
I only buy transformers that come with test results for how the secondary voltage behaves under load up to the rated current. One manufacturer in the USA that provides this is ANTEK, and their transformers are a good bargain IMO. With actual data in hand, you can make decisions and know exactly what you can expect from the hardware.
If you can't get data, then you should over-compensate (unless you want to buy one and measure it). For your case I would get get a good quality 100-150VA toroidal transformer. This is even more than the "twice amplifier power" rule (which I think is nonsense anyway) but it is because of the problems with most low VA transformers and NOT because 2x rated power is some magic formula. It's a very brainless metric IMHO.
For DIY use only, you can really bend the rules. For example, let's say that you want to build a higher power amp (e.g. 100W x2). It's possible to use LESS THAN the full amplifier power when buying the transformer if you will ONLY use the amp for music playback at modest levels with non compressed program material. This is because the AVERAGE power level is often only around a couple of watts and the 100W of amp output power are really only used to provide large reserves for peaks in the signal and to keep the amount of clipping to a minimum. In that case the capacitors are providing the instantaneous power (NOT the transformer!), and a transformer with a VA rating as low as HALF the full output power would be just fine. You read that right, HALF. This would be completely OK as long as you know that the amp is limited to operating at low average power and you only use it that way. No problem for DIY, but a commercial amp would NEVER be designed this way because they MUST be designed for abuse and worst case scenarios like "turn it up to 11" type use.
NOTE: I should add that I am assuming a class-AB amplifier here. Class-A is a completely different story.
Last edited:
Hi AndrewT, just wondering why a single ended supply and bridged topology is a bad idea for an active speaker? I work for Xkitz, we've had many of our customers build active speakers, and so far nothing but positive results.
Last edited:
Hey, and thank you so much for the feedback.
The transformer i chose because the manual for the PSU states: "You should select a transformer with a maximum secondary voltage of 26VAC, with a power rating of 50VA"
I might have missinterpreted that as max 50VA as well.
Single-ended supply - as in only positive rail?
Bridged topology - amp and x/o in same circuit?
Why does that sound terrible? I'm asking because I've only just started researching the subject and there is obviously a lot of different aspects to contemplate. The volume I have at disposal is very limited and I'm trying to keep down the size, so a combined circuit sounded good to me.. This speaker is not meant to be any kind of ultra hi-fi equipment, but rather a portable jamming device with synthezisers as feed. But I still want to make it sound decent and try to find some golden middle path here.
@Charlie: According to the manual for the amp module, it accepts 22V–50V DC, but I guess the PSU can be a bit cranky, considering the above statement about 26VAC max.
Looking at toroidal transformers in the VA range you suggest, the price really takes a leap up. I still have a lot of components to buy, and with the more powerful psu and a "proper" transformer kicks it up about $60. Right now I'm feeling a bit exhausted, so many components and dependencies to take in to account.
So.. What are my options here? Trying to find a linear AC/DC psu with that kind of voltage and power is not easy, and switchers I've heard 90% bad things about. Laptop chargers up to 120W I can get for free from scrapped computers at work, is there any way to clean up the signal enough for it to be usable? Does it help if i can get a class 1 with ground?
The transformer i chose because the manual for the PSU states: "You should select a transformer with a maximum secondary voltage of 26VAC, with a power rating of 50VA"
I might have missinterpreted that as max 50VA as well.
Single-ended supply - as in only positive rail?
Bridged topology - amp and x/o in same circuit?
Why does that sound terrible? I'm asking because I've only just started researching the subject and there is obviously a lot of different aspects to contemplate. The volume I have at disposal is very limited and I'm trying to keep down the size, so a combined circuit sounded good to me.. This speaker is not meant to be any kind of ultra hi-fi equipment, but rather a portable jamming device with synthezisers as feed. But I still want to make it sound decent and try to find some golden middle path here.
@Charlie: According to the manual for the amp module, it accepts 22V–50V DC, but I guess the PSU can be a bit cranky, considering the above statement about 26VAC max.
Looking at toroidal transformers in the VA range you suggest, the price really takes a leap up. I still have a lot of components to buy, and with the more powerful psu and a "proper" transformer kicks it up about $60. Right now I'm feeling a bit exhausted, so many components and dependencies to take in to account.
So.. What are my options here? Trying to find a linear AC/DC psu with that kind of voltage and power is not easy, and switchers I've heard 90% bad things about. Laptop chargers up to 120W I can get for free from scrapped computers at work, is there any way to clean up the signal enough for it to be usable? Does it help if i can get a class 1 with ground?
The datasheet is deceptively written but in any case those specs are overblown and there is no way you can get 40W into your biamped system, sorry.
1) all amp maker claims is:
40W into woofer + 40W into tweeter?
X watts into woofer, Y watts into tweeter, X+Y=40W?
And why use peak watts? .... if only a single rating scale is specified, honest one is RMS; all others, although mathematically correct, give "bigger numbers" to inflate ads and brochures.
2) in any case I checked TI LM4766 datasheet, http://www.ti.com/lit/ds/symlink/lm4766.pdf it´s a dual amp, may be used in stereo OR bridged OR biamped, not 2 or 3 options at once, just pick one.
It seems to be the biamped option.
In that case, with that power transformer, a bridge and a couple capacitors, you can have RAW +/-16V or +32V single supply.
You can realistically get some 8/10 W RMS into 8 ohms and twice as much into 4 ohms.
Fine for the tweeter, in fact way too much, not enough for the woofer.
3) IF you want something closer to rated power, datasheet shows it can put 40W RMS into 8 ohms but with +/-30V and probably +60V if single supply ; you have half that which means 1/4 the possible power.
4) IF you regulate what you have, you will have even lower power.
1) all amp maker claims is:
How do we read that?
40W into woofer + 40W into tweeter?
X watts into woofer, Y watts into tweeter, X+Y=40W?
And why use peak watts? .... if only a single rating scale is specified, honest one is RMS; all others, although mathematically correct, give "bigger numbers" to inflate ads and brochures.
2) in any case I checked TI LM4766 datasheet, http://www.ti.com/lit/ds/symlink/lm4766.pdf it´s a dual amp, may be used in stereo OR bridged OR biamped, not 2 or 3 options at once, just pick one.
It seems to be the biamped option.
In that case, with that power transformer, a bridge and a couple capacitors, you can have RAW +/-16V or +32V single supply.
You can realistically get some 8/10 W RMS into 8 ohms and twice as much into 4 ohms.
Fine for the tweeter, in fact way too much, not enough for the woofer.
3) IF you want something closer to rated power, datasheet shows it can put 40W RMS into 8 ohms but with +/-30V and probably +60V if single supply ; you have half that which means 1/4 the possible power.
4) IF you regulate what you have, you will have even lower power.
JMFahey, I really appreciate the breakdown. I was wondering about the wording as well, "produce 40W peak on both woofer and tweeter", if it was combined or per channel. But it looks like I should look for another amp for my setup.
I've asked this before in another thread, and I guess it's a bit off topic.. But if I were to run a 2-channel amp rated at say 60W RMS (that should be enough for this setup i think) as an BI-Amp, will it affect the amp in a bad way that the power draw from the tweeter and woofer are so different? The X/O will be at 4.5kHz..
I've asked this before in another thread, and I guess it's a bit off topic.. But if I were to run a 2-channel amp rated at say 60W RMS (that should be enough for this setup i think) as an BI-Amp, will it affect the amp in a bad way that the power draw from the tweeter and woofer are so different? The X/O will be at 4.5kHz..
It is 40W total between woofer and tweeter, same as the LM4766 datasheet.
The XAMP-M2 is is our lowest power bi-amplifier. It uses the two amps in the LM4766 to separately drive the woofer and tweeter, and it runs them in single ended configuration (like the 'single supply' reference design in fig 6 of the LM4766 datasheet). Our higher power version (XAMP-M3) uses one LM4766 for the woofer and one for the tweeter, and it runs the two amps in each of the chips in bridged configuration (each speaker terminal is driven by it's own amp, and the amps are run differentially).
BTW we use the term 'bi-amp' in the sense that we have an active crossover ahead of the amps and the woofer and tweeter are driven directly by the two amps.
Regarding the XAMP-M2 RMS power: we get a max of about 9V RMS out of each channel, which is about 10W into 8 ohms and 20W into 4 ohms. And that's with a 34V supply voltage.
+1
That´s honesty
As of the different power requirements for woofer and tweeter, they vary depending on crossover frequency ; at a quite high one such as 4500Hz program material fed to the tweeter will be minimal, while practically the whole load will be carried by the woofer/midrange.
Maybe if XKitz has some board with a bridged LM4766 for the woofer and, say, a TDA20xx or an LM1875 for the tweeter (with their versatile crossover included of course) then it might be a nice option.
If LF amp is bridged and HF one is a straight one, then even using the same power supply voltage, power difference is 4:1 , which looks more balanced given the probable use.
Or, using a split supply, say the 30+30V recommended by TI in the datasheet for full power out, the tweeter amp can be run as a single supply one, from ground to the + rail, again power difference will be 4:1 .
I commercially make guitar amps since forever, and a popular, battery powered one has a bridged TDA2005 driving a custom built 3 ohm speaker for almost 20W RMS plus a TDA2003 driving a LeSon 4 ohms Piezo tweeter (those have an internal autotransformer) for almost 10W ; all fed from a 12.6V 7A battery, sound is incredibly loud and clean for its diminute size, and highs are transparent and powerful.
We thought about the TDA20xx line for the tweeter channel on the XAMP-M3, but they couldn't take the supply voltage range we needed. In the end we decided to just stick with the bridged LM4766 for the tweeter channel. It is overkill, and a bit of a waste of potential power, but the gains are adjustable, so we can just dial back the tweeter power as needed.
I love that TDA based guitar amp. It's amazing how loud 20W can be without that passive XO in the way!
I love that TDA based guitar amp. It's amazing how loud 20W can be without that passive XO in the way!
I have decided to try out the cheap china boards with the TDA7498. I've read through the thread about them http://www.diyaudio.com/forums/class-d/265387-ebay-cheap-tda7498-boards-23.html and the people there seem really happy with them.
I ordered this one TDA7498 100W +100W High-power Digital Amplifier Board AMP Board with Radiator | eBay which seems to perform good, the inductors seem to be of higher quality than other similar builds.
The guys over there power these amps by laptop adapters or caged stationary variants. According to the thread 24V seems to be enough (well, some guys drive them with 19V adapters with good enough results), and in order to keep down temps I think that will have to be enough for me. I was going to order a 24V 5A, but I'd really like a 24V 3A because of the size difference.. I doubt i will get even close to 72W. But maybe I'm just being stupid here?
https://www.aliexpress.com/item/1-2-3-5-8-5-10-15-20-25-30A-LED-power-supply-transformer-AC/32662345920.html
Mind you that this is not a HiFi speaker, but merely something for when me and my friend twist some knobs in a basement (read that in american english please). But it's hard not to get sucked in to getting high quality stuff once you read up on everything. I have a feeling there is a more advanced project coming in the future (when i have space for a workshop and more time).
I ordered this one TDA7498 100W +100W High-power Digital Amplifier Board AMP Board with Radiator | eBay which seems to perform good, the inductors seem to be of higher quality than other similar builds.
The guys over there power these amps by laptop adapters or caged stationary variants. According to the thread 24V seems to be enough (well, some guys drive them with 19V adapters with good enough results), and in order to keep down temps I think that will have to be enough for me. I was going to order a 24V 5A, but I'd really like a 24V 3A because of the size difference.. I doubt i will get even close to 72W. But maybe I'm just being stupid here?
https://www.aliexpress.com/item/1-2-3-5-8-5-10-15-20-25-30A-LED-power-supply-transformer-AC/32662345920.html
Mind you that this is not a HiFi speaker, but merely something for when me and my friend twist some knobs in a basement (read that in american english please
). But it's hard not to get sucked in to getting high quality stuff once you read up on everything. I have a feeling there is a more advanced project coming in the future (when i have space for a workshop and more time).It would be a good idea to order a dozen spare tweeters. with this amount of power it is easy to destroy the tweeter. I have not seen many that can really handle 15W rms. sometimes the tweeter rationg is given as ""suitable for a system power of xx watts"".
you may want to consider some passive protection like a ptc or lightbulb.
you may want to consider some passive protection like a ptc or lightbulb.
I think Xkitz's line of bi-amplifiers are EXCELLENT ideas: affordable biamping with integral L-R 24db/octave crossovers. The newest one ([X-AMP-M4]) is truly a bargain. However, I also would like to see only true RMS ratings used, with power supply required! This is obviously intended for the DIY community, and we KNOW the difference between peak and RMS---don't try to 'fool' us with an inflated spec!
Tweeter protection is definitely a good idea. I work for the supplier of the XAMP line, and on those the power to the tweeter (and the woofer) is adjustable with a trim-pot, so you can always dial it back to a safe level. They also have a mute circuit that kicks in when the power is turned on or off, so there's no power on/off spike that could blow a tweeter. But that's a very good idea, just for extra protection, I think I'm going to lobby to get a PTC added to the tweeter channel on the XAMP boards.
Thanks for the tips, that PTC is news to me! I don't know that much about ribbon tweeters, but looking at the spec sheet, it has a flat 8 Ohm impedance response. Not sure if that changes anything.. I will run my crossover at 4.5kHz, so the power to the tweeter is going to be fairly low I think.
But the PTC is a cheap insurance, so I will definitely get one. But, looking at the calc in the link provided:
- tweeters with 10 mm dome diameter ca. 1 Watt should be entered
- tweeters with 19 mm dome diameter ca. 4 Watt should be entered
- tweeters with 25 mm dome diameter ca. 8 Watt should be entered
- PA tweeters half the the rated power handling should be entered
It's not a dome and I don't know if a ribbon is more fragile or less than a dome tweeter.
But the PTC is a cheap insurance, so I will definitely get one. But, looking at the calc in the link provided:
- tweeters with 10 mm dome diameter ca. 1 Watt should be entered
- tweeters with 19 mm dome diameter ca. 4 Watt should be entered
- tweeters with 25 mm dome diameter ca. 8 Watt should be entered
- PA tweeters half the the rated power handling should be entered
It's not a dome and I don't know if a ribbon is more fragile or less than a dome tweeter.
PTCs aren't very good for sound quality.
Ribbon tweeters should always have a large series cap, both for protection of the ribbon tweeter, and to counteract a near zero dc resistance.
As for a candidate for active amp, Hypex is doing end of production sale on UcD32 and UcD34 modules. 2 and 4 channels class D with power supply for €50 and €65, respectively. Note that you'll need an input buffer on these, and they don't come with a crossover.
Johan-Kr
Ribbon tweeters should always have a large series cap, both for protection of the ribbon tweeter, and to counteract a near zero dc resistance.
As for a candidate for active amp, Hypex is doing end of production sale on UcD32 and UcD34 modules. 2 and 4 channels class D with power supply for €50 and €65, respectively. Note that you'll need an input buffer on these, and they don't come with a crossover.
Johan-Kr
From Wiki:
When power is removed, the heating due to the holding current will stop and the PPTC device will cool. As the device cools, it regains its original crystalline structure and returns to a low resistance state where it can hold the current as specified for the device. This cooling usually takes a few seconds, though a tripped device will retain a slightly higher resistance for hours, slowly approaching the initial resistance value. The resetting will often not take place even if the fault alone has been removed with the power still flowing as the operating current may be above the holding current of the PPTC. The device may not return to its original resistance value; it will most likely stabilize at a significantly higher resistance (up to 4 times initial value). It could take hours, days, weeks or even years for the device to return to a resistance value similar to its original value, if at all.[1] Since a PPTC device has an inherently higher resistance than a metallic fuse or circuit breaker at ambient temperature, it may be difficult or impossible to use in circuits that cannot tolerate significant reductions in operating voltage, forcing the engineer to choose the latter in a design.
When power is removed, the heating due to the holding current will stop and the PPTC device will cool. As the device cools, it regains its original crystalline structure and returns to a low resistance state where it can hold the current as specified for the device. This cooling usually takes a few seconds, though a tripped device will retain a slightly higher resistance for hours, slowly approaching the initial resistance value. The resetting will often not take place even if the fault alone has been removed with the power still flowing as the operating current may be above the holding current of the PPTC. The device may not return to its original resistance value; it will most likely stabilize at a significantly higher resistance (up to 4 times initial value). It could take hours, days, weeks or even years for the device to return to a resistance value similar to its original value, if at all.[1] Since a PPTC device has an inherently higher resistance than a metallic fuse or circuit breaker at ambient temperature, it may be difficult or impossible to use in circuits that cannot tolerate significant reductions in operating voltage, forcing the engineer to choose the latter in a design.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.