There is no idle voltage issue - it's just that big iron is expensive. I would just add a bunch of big 10000uF caps and a bridge rectifier. This is the "ideal" way the more expensive high end amps provide their dual rail voltage. I think the IRS2092 is expecting between +/- 30v to 60v with recommended voltage of +/-50v These Antek's are nice - usually the prices are closer to $120 for this trafo. Since these are AC ratings I think divide by 0.707 to get the peak full wave bridge rectified rail voltage. So a 35v unit would yield 50v rails.
http://www.antekinc.com/an-4435-400va-35v-transformer/
Last edited:
400Hz and 4-5kHz?
LOL, you've nearly described a classic 70s/80s 3 way. Still my preferred format. In fact this is what's planned for my next audio project, using some FE108EZ out of my stash for the mids and peerless 10" for woofers. I guess I'll have to discuss in on the multi-way forum. IME 4-5k about ideal. If this isn't working I like to go to a true supertweeter crossed 10k or higher. A bad crossover at 6k is just as audible to me as one at 4k, and the wavelengths are shorter.
Waveguides weren't usually used for the 70's monkey boxes -- with some exceptions, but a cone tweeter in some ways is better for matching dispersion to the mid than a dome. Many people preferred the original cone tweeter on JBL L100s partly for this reason.
400Hz and 4-5kHz?
IME 4-5k about ideal. If this isn't working I like to go to a true supertweeter crossed 10k or higher. A bad crossover at 6k is just as audible to me as one at 4k, and the wavelengths are shorter.
Waveguides weren't usually used for the 70's monkey boxes -- with some exceptions, but a cone tweeter in some ways is better for matching dispersion to the mid than a dome. Many people preferred the original cone tweeter on JBL L100s partly for this reason.
Well... I only said 4-5K because X is really concerned about minimal xo phase shifts in the telephone band and what not. I usually worry a bit more about power response and lobbing, but I'm not necessarily of the crowd that always has to cross a tweeter as low as possible. That being said, the 10f is so small and well behaved that maybe I would cross it higher than I normally would, maybe as high as 6 or 7K.
A waveguide is certainly not needed, but it could be used to help match dispersion characteristics. Or a cone tweeter could probably do the same thing, but the 10f is already almost a cone tweeter so that may be a little redundant.
Hi X, it is hard to compare sounds with non-identical music, but from memory I can say that this amp is much more fatiguing than your previous one. Of course the bass is more dynamic here, but my preference is always related to distortion.
Nothing can change that, except if there is a way to lower the feedback gain.
But I'm sure that you can fix the low frequency performance (and use amp with low THD) if you can make something such that the 10F will roll-off at similar speed with RS225. Yours is similar in less than one octave below crossover.
I think the default gain set on this amp is kind of high at 32dB vs 26dB for the TPA3116. The frequency response measures just a bit hotter (0.5dB) in the high frequency range 10k and above. Not sure if that is enough to make it fatiguing. It may just be the source material. Oingo Boingo is full of artificial synth guitar.
Ah, 32dB vs 26dB. That's enough for me to explain the difference in distortion perception (which favors lower gain) and dynamic (which favors higher gain). But FAST is naturally very dynamic and easy to drive so high gain is not necessary (cheap amps use excessive gain simply to increase power).
I did a more careful measurement of the response with the TDA7492, it does have a slightly hotter top end. I tried applying -1.5dB high shelf from 3kHz at Q=0.3 and that evened out the response. It sounds less fatiguing now but detail in highs is still more clear than 3116 and bass is definitely more punchy. I am kind of surprised how good it sounds actually. It's not just the higher gain because I attenuated the input gain from miniDSP to compensate. The sound is just more dynamic.
I just ordered the 100watt big brother of this amp, the TDA7498 for $11. Let's see how the woofers work with that guy. More info in the TDA7492 Amp thread.
http://www.diyaudio.com/forums/class-d/275505-tda7492-amp-2.html
I just ordered the 100watt big brother of this amp, the TDA7498 for $11. Let's see how the woofers work with that guy. More info in the TDA7492 Amp thread.
http://www.diyaudio.com/forums/class-d/275505-tda7492-amp-2.html
There is nothing to build. Just connect DC power supply and speaker leads. Connect audio input. Done. Well if you want a box then that may be considered building an amp.
If you want to upgrade caps etc then there is a little work but nothing someone familiar with a soldering iron can't do.
I find that either the TPA3116 or TDA7492 have plenty of power with a 19v laptop brick.
If you want to upgrade caps etc then there is a little work but nothing someone familiar with a soldering iron can't do.
I find that either the TPA3116 or TDA7492 have plenty of power with a 19v laptop brick.
What amp would you pick if you're looking for 250 watt for each channel? Two amps, one for each channel would be alright too . Minimising the wire length as each amp would go directly behind the speaker.
I bet that would need a pretty good power supply as well. Hmmm... not ready yet as I am broke.
. Minimising the wire length as each amp would go directly behind the speaker.I bet that would need a pretty good power supply as well. Hmmm... not ready yet as I am broke.
What amp would you pick if you're looking for 250 watt for each channel? Two amps, one for each channel would be alright too
I bet that would need a pretty good power supply as well. Hmmm... not ready yet as I am broke.
The Sure IRS2092 boards from ebay or Parts Express have that kind of guts, reviews are mostly pretty good, but there seem to be some complaints about quality and consistency
For just a little more Hifimediy or Connexelectronic are probably better quality and have better product support
Hifimediy T3S mcap
Hifimediy T4 mcap
Or maybe one of the IRS2092 amps from Connexelectronic:
Connexelectronic
Of course you could also do what I did when I was thinking about building a couple of these and instead get a crown XLS1000 or 1500. I bought a couple of XLS1500 in like new condition from Guitar Center for $225 each. Outstanding sound quality and they even include fully transferable warranty.
I also picked up an Inuke3000 new off ebay for a great price. This is decent if you want dsp too.
Connexelectronic also makes nice SMPS dual rail supplies that these kind of amps need. Good for LM386 amps too.
Connexelectronic
You are right though about the Crowns: not a bad price considering it has smps, case, jacks, etc and a warranty.
Connexelectronic
You are right though about the Crowns: not a bad price considering it has smps, case, jacks, etc and a warranty.
Yes, their power supplies look really nice. I was considering them at one time
That's the thing that swayed me. I was going to build a couple of these boards for a dipole project, one class D and one T amp, but after figuring the cost of the case, power supply, boards and all the odds and ends, it was pretty much cheaper to get the Crowns. And I really like the fact that they're pretty much bullet proof, I Run one channel of the Inuke at 2 ohms with no worries.
One big consideration with pro amps is the fan noise and input sensitivity. The Inuke sucks for fan noise even after a fan mod, but the crowns are silent. The crown amps also have binding post and rca's which is a plus.
Guitar Center has a used section online that you have to check often to get what you want at the price you want to pay; there's a lot of variation on pricing of the used items between stores.
That's the thing that swayed me. I was going to build a couple of these boards for a dipole project, one class D and one T amp, but after figuring the cost of the case, power supply, boards and all the odds and ends, it was pretty much cheaper to get the Crowns. And I really like the fact that they're pretty much bullet proof, I Run one channel of the Inuke at 2 ohms with no worries.
One big consideration with pro amps is the fan noise and input sensitivity. The Inuke sucks for fan noise even after a fan mod, but the crowns are silent. The crown amps also have binding post and rca's which is a plus.
Guitar Center has a used section online that you have to check often to get what you want at the price you want to pay; there's a lot of variation on pricing of the used items between stores.
In the sub 50w range the TPA3116's and TDA749x's seem very cost effective vs the bigger pro amps. I know that the sound quality of a modified TPA3116D2 is better than a Crown XLS or Alesis rack mount amps if operating within the range of the smaller TPA3116D2. I sent a modded (OSCON's, bootstrap snubbers, 3.3uF film input caps, CoilCraft chokes) YJ Blue/black 3116 board to a friend of mine who has several pro amps and he tested them and said the sound quality of the small amp is better. This may make sense to use the smaller ones for the full range/tweeter and the bigger pro amp for the woofer or sub where you need the power. But I would not count on a big rack amp to power my HF section for best sound.
This may or may not be true. I don't doubt that they sound good. One thing to consider is that 1) not all pro amps sound the same, just like not all Hifi amps do and 2) the XLS1500 and up is better sounding than the XLS1000 especially in the mid range and treble because of higher SR.
My pro amps replaced the amp section in my $1300 Peachtree Nova and with the Peachtree Pre and DAC feeding the crowns, the sound is unquestionably better, hands down. Clearer, cleaner, more solid, more open and more neutral.
So, I'm sure that your amps sound good, but the opinion of one guy, in one room, with one set of equipment is not necessarily fact, just like my opinion on the matter.
One thing that is a fact is if you don't have enough power, and really a good chunk of headroom, the sound will suffer. Running an 8" in a small sealed box with an LT circuit on maybe 50w doesn't not sound like enough to me. Have you modeled this in a program like Winisd to see what kind load this is putting on your amps and the driver?
Agreed that not all amps are the same and I did not realize the Crown XLS1500 has better SQ than the 1000, although it was the 1500 that was used. I have a problem using WinISD so I need to figure out how to put the same bi-quad coefs that I use for the LT in miniDSP into the bi quad filter section of Akabak and then check the model there.
When I cranked volume up to max with my current 50w amp, visually it seems like I am able to hit 7mm xmax. So I may not really need all this extra amp power.
When I cranked volume up to max with my current 50w amp, visually it seems like I am able to hit 7mm xmax. So I may not really need all this extra amp power.
I'm sure you don't need 250W, but headroom never hurts, especially when boosting the low end. 50W is a solid number in my opinion as long as it can do that with low distortion, but sealed is somewhat inefficient and an LT is a power hog depending on the native response and the desired f3.
I took a look at a Winisd model with the RS225-04 driver that I already had saved. In a 26L box an LT providing an f3@30Hz requires 100W@30Hz and 200W@20 for about 94db after baffle step correction. This is with 20W input power.
I'm sure this is totally different than you have and maybe yours isn't this aggressive, but you can see the demands that can be put on the amp.