-hello, i am building a triska sub for my co-worker (link below). i have the sub finished down to mounting drivers and passives. i am very interested in building the plate amp for him as i have everything i need laying around. the dayton 8"DVC is listed @ 2x8ohm and a RE of 3ohm.
from what i have read i would need a parrallel amp to handle the low ohm driver. if it was mine id go all out and do a BPA for the learning experience (in the true DIYAUDIO spirit). however, im doing this free of charge, and i would like to build what i need to drive it properly.
the heatsink i plan to use would be overkill for any gainclone i have built in the past. i would like to know if i went bridged with a drv134 would it be too much to ask of 2Xopa549? to drive a small 4ohm sub?
i plan to run the chipamps under its max voltage @25 to 28 volts+- and prolly a gain around 15 or so. i dont have any 1% 0.1-0.3ohm power resistors and prolly dont want to match anything by hand either. i would think that not using the max voltage and heavy heatsinking would maybe keep the chip safe thermal wise?any body care to show me the light?
this guy has your standard computer speaker system and plays online games.i am trying to build him a nice stereo sytem for music- i am sure this sub/sat system will be a upgrade of many levels
and he prolly wont run it up near its maxx. however, i do not want to set my self up to re-build or upgrade in the future either.
TIA , steve
*sorry for bad pics, this is what your digital camera does after your 3yo dumps your 1yo bottle into your camera!*
http://www.partsexpress.com/projectshowcase/triska/driver.html
from what i have read i would need a parrallel amp to handle the low ohm driver. if it was mine id go all out and do a BPA for the learning experience (in the true DIYAUDIO spirit). however, im doing this free of charge, and i would like to build what i need to drive it properly.
the heatsink i plan to use would be overkill for any gainclone i have built in the past. i would like to know if i went bridged with a drv134 would it be too much to ask of 2Xopa549? to drive a small 4ohm sub?
i plan to run the chipamps under its max voltage @25 to 28 volts+- and prolly a gain around 15 or so. i dont have any 1% 0.1-0.3ohm power resistors and prolly dont want to match anything by hand either. i would think that not using the max voltage and heavy heatsinking would maybe keep the chip safe thermal wise?any body care to show me the light?
this guy has your standard computer speaker system and plays online games.i am trying to build him a nice stereo sytem for music- i am sure this sub/sat system will be a upgrade of many levels
and he prolly wont run it up near its maxx. however, i do not want to set my self up to re-build or upgrade in the future either.TIA , steve
*sorry for bad pics, this is what your digital camera does after your 3yo dumps your 1yo bottle into your camera!*
http://www.partsexpress.com/projectshowcase/triska/driver.html
Attachments
Nice size heat sink. If I understand correctly, you have a 8" dual voice coil driver that is rated at 8 ohms per VC. Your supply voltage will be around +/-25V and peak out at about +/-28V. If this is all correct there is nothing to worry about thermally. You have a few options.
1) Use 2xLM3886 and one VC from each chip, each chip sees 8 ohms, no issues.
2) Wire VC in parallel and run from one chip, will see 4 ohms, no issues again.
Going for a bridge design will not work out very well because then you will need either 4 chips or wire VC in series but that won't do anything compared to driving each VC with a separate chip. The LM3886 will not work well if it sees less than 4 ohms rated load. So bridge into VC wired in parallel will mean each LM3886 will see 2 ohms and that is bad. No matter what you do you will not get more power than the other options. Running parallel into VC wired in parallel (4ohms) will not get more power either. Each chip will see 8 ohms so same as #1 above, really.
Using the Overture Design Guide (get it at: http://www.national.com/appinfo/audio/) your 3 options above with +/-25V supply have the following output powers.
1) 30W per VC, 60W total in to sub driver.
2) 58W into sub driver
Both are about the same thermally too if both chips are on the same heat sink. So really there is not much difference between the two options except cost and complexity.
-SL
1) Use 2xLM3886 and one VC from each chip, each chip sees 8 ohms, no issues.
2) Wire VC in parallel and run from one chip, will see 4 ohms, no issues again.
Going for a bridge design will not work out very well because then you will need either 4 chips or wire VC in series but that won't do anything compared to driving each VC with a separate chip. The LM3886 will not work well if it sees less than 4 ohms rated load. So bridge into VC wired in parallel will mean each LM3886 will see 2 ohms and that is bad. No matter what you do you will not get more power than the other options. Running parallel into VC wired in parallel (4ohms) will not get more power either. Each chip will see 8 ohms so same as #1 above, really.
Using the Overture Design Guide (get it at: http://www.national.com/appinfo/audio/) your 3 options above with +/-25V supply have the following output powers.
1) 30W per VC, 60W total in to sub driver.
2) 58W into sub driver
Both are about the same thermally too if both chips are on the same heat sink. So really there is not much difference between the two options except cost and complexity.
-SL
BTW, I think that heat sink is over kill for this project. If you really want more power than get a different supply and run it more like +/-35V and use 2xLM3886, one wired to each VC and you will get more like 50-60X per VC so 100 - 120W into the driver. Then that heat sink might be better used.
-SL
-SL
well i only have 1 lm3886 hanging around so im using opa549, which i like for midbass/lows anyways.
that being said i have several transformers pushing over 30v AC but then id might be way to high for these chips max volts +-30 IIRC
so you say drive both chips into a voice coil? and dont drive it bridged with the sub wired 4ohms. i have always done the latter, but im a car stereo guy learning about amps
i am trying to keep this project confined to my spare parts bin if you can understand
*about the heatsink, it is overkill! but it is junk also. pulled a stack of them out of the dump. **see pic**
i am stacking these and using them for my son of zen amp 32 7.5ohm vishay rh-50 resistors. and 2 more sinks for each of the 2 sets of mosfets.
that being said i have several transformers pushing over 30v AC but then id might be way to high for these chips max volts +-30 IIRC
so you say drive both chips into a voice coil? and dont drive it bridged with the sub wired 4ohms. i have always done the latter, but im a car stereo guy learning about amps
i am trying to keep this project confined to my spare parts bin if you can understand
*about the heatsink, it is overkill! but it is junk also. pulled a stack of them out of the dump. **see pic**
i am stacking these and using them for my son of zen amp 32 7.5ohm vishay rh-50 resistors. and 2 more sinks for each of the 2 sets of mosfets.
Attachments
If you only have 1 LM3886 then tie the VC in parallel and run it from the tansformer you have. At 4 ohms you want the supply to be around +/-28V, give or take. The maximum voltage a LM3886 can handle is +/-42V or with no signal +/-47V (per the datasheet). But when driving 4 ohms you need to stay below +/-30V or it will over heat, no matter what heat sink you use. For music it will probably be OK but I typically prefer to be conservative. At +/-30V you will get over 80W so that is a significant amount of power for a little 8" driver.
I missed that you had one chip of each to use. It looks like the OPA549 is limited to a maximum supply of +/-30V. So I think the best thing is to get a supply aroun d+/.-30V and run just the LM3886 into the driver with the VC in parallel, as explained above.
At BTL you are driving 2 ohms per IC, too low. At parallel that is the same as driving each VC with it's own IC, nothing to gain here. IF the OPA549 was not limited to no higher than +/-30V I'd say use them both, each driving a single VC and run the supply like +/-35V or a tad higher. That won't work with the limits on the OPA chip. If you think through the possibilities you will probably see there isn't much else to do.
Godd luck,
-SL
I missed that you had one chip of each to use. It looks like the OPA549 is limited to a maximum supply of +/-30V. So I think the best thing is to get a supply aroun d+/.-30V and run just the LM3886 into the driver with the VC in parallel, as explained above.
At BTL you are driving 2 ohms per IC, too low. At parallel that is the same as driving each VC with it's own IC, nothing to gain here. IF the OPA549 was not limited to no higher than +/-30V I'd say use them both, each driving a single VC and run the supply like +/-35V or a tad higher. That won't work with the limits on the OPA chip. If you think through the possibilities you will probably see there isn't much else to do.
Godd luck,
-SL
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.