Would it be possible to bridge two modules? If its possible, would you recommend against it for any reason?
I'm sure it sounds just as good or better than all the different t-amps, but I think a step down in power from a t-amps already modest ouptut would be a hard hit for anyone that doesn't have extremely efficient speakers.
I'm sure it sounds just as good or better than all the different t-amps, but I think a step down in power from a t-amps already modest ouptut would be a hard hit for anyone that doesn't have extremely efficient speakers.
Hi anonymous1. It probably is possible to bridge two modules but I haven't tried that yet. I will when I get the my two testers built up.
The D10.1s have just as much power output capability as any Tripath amp running off of a 12V rail. It's the TA2021B and it's ability to run off of 14V that is capable of a little more power output.
Oh and the second kit is spoken for 🙂
The D10.1s have just as much power output capability as any Tripath amp running off of a 12V rail. It's the TA2021B and it's ability to run off of 14V that is capable of a little more power output.
Oh and the second kit is spoken for 🙂
Sorry, I should have been more clear. My statement was based on my personal experience of running a t-amp at <14v as apposed to 12v. IMHO, that 2v makes a noticable difference, especially in the bass and mid-bass.
I think if you already have taken the extra step to use a +/- supply to get the most out of this chip, you might as well go bridged so you can take home the trophy in a D10.1 vs. T-amp battle.
I think if you already have taken the extra step to use a +/- supply to get the most out of this chip, you might as well go bridged so you can take home the trophy in a D10.1 vs. T-amp battle.
Hi,
I'm an electronic newbie but I've already built an amp6 and since I'm looking for a good solution in a 5.0 DIY amp, your mono modules seems very interesting to me.
I would have liked to test your new kit but:
1/ I'm going on vacation for 3 weeks in August
2/ I don't own any +12V -12V power supply
Keep "feeding us" with your results. 🙂
I'm an electronic newbie but I've already built an amp6 and since I'm looking for a good solution in a 5.0 DIY amp, your mono modules seems very interesting to me.
I would have liked to test your new kit but:
1/ I'm going on vacation for 3 weeks in August
2/ I don't own any +12V -12V power supply
Keep "feeding us" with your results. 🙂
theAnonymous1 said:
Assuming the Tripath chip can supply adequate current, the maximum output with a 14V rail would be a 14Vp-p sine wave. With an 8ohm load that's 12.25Wrms. With a 4ohm load that's 24.5Wrms.
With a 12V rail the maximum power drops to 9Wrms at 8ohms and 18Wrms at 4ohms.
So that's a 3.25W at 8ohms and 6.5W at 4ohms difference in power output. The extra bit of headroom is nice to have but I haven't been able to notice a difference in sound quality between a 12V rail and the 14V rail I run my TA2021B amps at. Just like the Tripath chips the MP7720 could be run with higher rail voltages but I don't know how it will handle it. I can up the rail voltages to +/-14V and see what happens when I get my tester module built.
I've got two kits ready to go to the two people in line, but am still waiting to get the chips in...
A question of power
Monolithic Power offers a dual channel version of this chip, the MP7722 that would be much more suited to doing a bridge tied load amp with. The problem is that the load would be limited to 8 ohms for various reasons including heat and distortion. There is also the question of switching frequency synchronization that would need to be answered.
They have a larger chip already out for driving bridged loads that will go down to 6 ohms. This MP7782 would be a better choice as it is rated up to 50 watts out. With the distortion characteristics almost flat with frequency this amp has a chance to be a winner but a lot of work would need to be done. This chip is a single channel so a stereo amp requires 2. With board layout and heatsinking it starts to look as if the amp one would be a better choice for now.
Roger
Monolithic Power offers a dual channel version of this chip, the MP7722 that would be much more suited to doing a bridge tied load amp with. The problem is that the load would be limited to 8 ohms for various reasons including heat and distortion. There is also the question of switching frequency synchronization that would need to be answered.
They have a larger chip already out for driving bridged loads that will go down to 6 ohms. This MP7782 would be a better choice as it is rated up to 50 watts out. With the distortion characteristics almost flat with frequency this amp has a chance to be a winner but a lot of work would need to be done. This chip is a single channel so a stereo amp requires 2. With board layout and heatsinking it starts to look as if the amp one would be a better choice for now.
Roger
Re: A question of power
The 7722 is basically just two 7720s on one chip. There doesn't seem to be much advantage to this since the 7722 comes in a TSSOP-20 package that is physically larger than 2 separate SOIC-8 packages. Maybe it costs slightly less than two SOIC-8 MP7720's...
Synchronizing the switching frequencies of each individual module is actually very simple. In fact, MPS has shown how to do this in the data sheet for the MP7731. Basically, they feed the audio signal to the inverting input of one amp and feed the output signal of that amp to the inverting input of the second amp. That synchronizes the switching and gives you the inverted audio signal needed for bridged operation.
I saw this the other week and intend to try it out with my D10.1 modules.
sx881663 said:
The 7722 is basically just two 7720s on one chip. There doesn't seem to be much advantage to this since the 7722 comes in a TSSOP-20 package that is physically larger than 2 separate SOIC-8 packages. Maybe it costs slightly less than two SOIC-8 MP7720's...
Synchronizing the switching frequencies of each individual module is actually very simple. In fact, MPS has shown how to do this in the data sheet for the MP7731. Basically, they feed the audio signal to the inverting input of one amp and feed the output signal of that amp to the inverting input of the second amp. That synchronizes the switching and gives you the inverted audio signal needed for bridged operation.
I saw this the other week and intend to try it out with my D10.1 modules.
Re: Re: A question of power
Allow me to correct myself... The TSSOP-20 package is actually only 1.6mm longer than a single SOIC-8 package because the lead pitch is smaller.
BWRX said:
Allow me to correct myself... The TSSOP-20 package is actually only 1.6mm longer than a single SOIC-8 package because the lead pitch is smaller.
BWRX said:
The Tripath chips I've worked with will not go rail to rail. Best I've seen is the 2020 which goes to about 1V under rail before clipping.
So for max power you should subtract a least 1 volt from rail. Maybe 2 volts for the SMD chips. That will certainly cut into your 12V rail and output power.
I agree and have been looking at the MP7782. I wonder if we can talk Brian into designing a PCB for it? 😉
Looks like it's time for me to bust out the 8ohm power resistors and see what output voltage the 7720 can reach before clipping with regulated 12V rails.
Talk me into making a board for the 7782? I've got enough stuff to keep me busy as it is but why not add one more project to the list
Talk me into making a board for the 7782? I've got enough stuff to keep me busy as it is but why not add one more project to the list
With the Ref-T and it's on board regulated 14V supply I saw a peak amplitude of approximately 12V at the onset of clipping into an 8ohm load.
Wit the D10.1s and regulated +/-12V rails I saw a peak amplitude of approximately 10.5V at the onset of clipping into an 8ohm load.
The test material was a 1kHz sine wave generated by my computer.
It's worth noting that the 7720 chip only got warm to the touch while the TA2021B got warm enough for me to feel the need to put a heatsink on the chip. Both modules amplified the test tone into the load for 1 minute at full output power without problems.
Wit the D10.1s and regulated +/-12V rails I saw a peak amplitude of approximately 10.5V at the onset of clipping into an 8ohm load.
The test material was a 1kHz sine wave generated by my computer.
It's worth noting that the 7720 chip only got warm to the touch while the TA2021B got warm enough for me to feel the need to put a heatsink on the chip. Both modules amplified the test tone into the load for 1 minute at full output power without problems.
Cool Brian, that fits with what I have found.
Odd that the Tripath gets a little closer to rail, and the 7720 runs cooler.
Of course the 7720 was supplying a little under 7 WPC as opposed to the AMP3's 9 WPC. A slight difference there.
Did the 7720 "soft clip" like the Tripath?
Odd that the Tripath gets a little closer to rail, and the 7720 runs cooler.
Of course the 7720 was supplying a little under 7 WPC as opposed to the AMP3's 9 WPC. A slight difference there.
Did the 7720 "soft clip" like the Tripath?
The TA2020, TA2021B, and TA2024 all have bridged outputs.
Each output floats at about half the supply voltage (6V) and then they swing in opposite directions (one goes to 12V while the other goes to 0V), theoretically giving you up to the full rail voltage across the load.
The 7720 is single ended and can theoretically swing up to the positive rail voltage (12V) or lower rail voltage (-12V) across the load because one side of the load is always connected to ground (0V). They both get you the same output power but a bridged amplifier only requires one supply rail while a single ended amp requires a split supply rail with a ground reference.
I'll try to get some scope shots of the outputs later this week.
Each output floats at about half the supply voltage (6V) and then they swing in opposite directions (one goes to 12V while the other goes to 0V), theoretically giving you up to the full rail voltage across the load.
The 7720 is single ended and can theoretically swing up to the positive rail voltage (12V) or lower rail voltage (-12V) across the load because one side of the load is always connected to ground (0V). They both get you the same output power but a bridged amplifier only requires one supply rail while a single ended amp requires a split supply rail with a ground reference.
I'll try to get some scope shots of the outputs later this week.
Interesting story involving my new computer and the D10.1 modules...
I just recently sold my old computer (Apple PowerBook G4) and used those funds to purchase a new MacBook Pro (it is very nice 🙂 ). I've used the Powerbok G4 as a source for the D10.1s and had no problems at all. The new MacBook Pro works fine with the D10.1s too, BUT something not so cool happens when the computer goes to sleep...
Something has to be going on with the audio out on the computer that causes the amps to emit an awful tone. I have the audio out of my computer going into a simple buffered preamp with gain that allows me to adjust volume of the input signal to the amps but this has worked fine and was also used with the PowerBook G4.
Anyway, I've had this awful tone incident happen on three occasions now and the chips did not survive this last one. The internal low side mosfet is shorted on both modules, so the chip needs to be replaced, but everything else appears fine.
I'll hook up my Ref-T and see if I can get the same thing to happen while I replace the chips on my two modules. It looks like some form of protection needs to implemented that disables the chip when the rail voltages deviate much. Or a DC blocking cap on the input might need to be used if this was caused by DC output from my computer or preamp.
I just recently sold my old computer (Apple PowerBook G4) and used those funds to purchase a new MacBook Pro (it is very nice 🙂 ). I've used the Powerbok G4 as a source for the D10.1s and had no problems at all. The new MacBook Pro works fine with the D10.1s too, BUT something not so cool happens when the computer goes to sleep...
Something has to be going on with the audio out on the computer that causes the amps to emit an awful tone. I have the audio out of my computer going into a simple buffered preamp with gain that allows me to adjust volume of the input signal to the amps but this has worked fine and was also used with the PowerBook G4.
Anyway, I've had this awful tone incident happen on three occasions now and the chips did not survive this last one. The internal low side mosfet is shorted on both modules, so the chip needs to be replaced, but everything else appears fine.
I'll hook up my Ref-T and see if I can get the same thing to happen while I replace the chips on my two modules. It looks like some form of protection needs to implemented that disables the chip when the rail voltages deviate much. Or a DC blocking cap on the input might need to be used
if this was caused by DC output from my computer or preamp.
I have also had my share of woes using a PC as a source; a dead T-amp, ground loops, power on/off noise etc.. I have a USB soundcard that emits a nasty sound when the PC is off.
I found the only good way around these problems is by using optical out to a decent DAC. Even an entry level DAC with a single TDA1543 sounds better than most PC outputs anyways.
I found the only good way around these problems is by using optical out to a decent DAC. Even an entry level DAC with a single TDA1543 sounds better than most PC outputs anyways.
My MacBook has an analog and optical audio out 🙂 You need a special cable to access the optical out though...theAnonymous1 said:
THe keywords in that sentence are "most PC outputs". I've found Macs to have very decent analog audio outputs.theAnonymous1 said:
theAnonymous1 said:
Yes it was 😉
In other news, I have successfully bridged two D10.1 modules! That's right, they're capable of being bridged and it's actually very simple to do so. All you need is a couple of resistors (and some alligator clips leads if you want to jerry rig it like I did).
There is definitely more power available as one would expect from a bridged configuration. I couldn't turn up the volume the whole way to see where the waveform clipped because it was too loud being within a few feet of my 90db 1W/1m speakers 🙂 I'll hook up my power resistors, find another scope probe, and see what kind of votlage swing the bridged modules are capable of.
They still sound very good, although there is a more noise... but most of that could be due to the feedback resistors right on top of the output wires, the alligator clip connections, and a ground loop
Attachments
- Status
- Not open for further replies.