if the middle section is separate from the rest, then easy enough to have a couple of vent holes... have some top and bottom and the hot air will naturally flow out the top and cool air in the bottom. or add a little cpu fan to extract the hot air. or do the diy plate amp thing. plenty of options.
Tiny practical problem with that arrangement. The entire top is a large 15Watt solar panel. So you can't have holes in the top. The bottom is recommend to place directly on ground or the tuning will be off (note to get correct tuning indoors you put it on stands at least 1m off the ground and with at least 40cm to the backwall) so holes in the bottom will not produce any useful resuslt. There are cooling holes already built-in as you can see on the picture but the air circulation is minimal. But as I said you could have a cooling fan that is either termally controlled or just kicks in whenever the voltage rises above 13.5V, ie. either form of charging is in progress.
I didn't include a fan because I don't think it's crucial and the lowest power fan I could find with suitable control circuitry uses about a 1.3 watts of power which is about a third of what the amp uses continously at maximum power, so adding one limit battery time too severely for my taste.
I didn't include a fan because I don't think it's crucial and the lowest power fan I could find with suitable control circuitry uses about a 1.3 watts of power which is about a third of what the amp uses continously at maximum power, so adding one limit battery time too severely for my taste.
ahh ok. in the photos it shows solar panels just on the ends.
1.3 watts is 1/3 of what the amp uses at max power? that doesnt sound quite right... you must be using more than 3.9W! the amp6 basic is 25W per channel for a total max power of 50W. Tim is using an Amp9, for up to 4x100W per channel - i know he probably wont use a fraction of that power, but even so i think some cooling arrangement is wise.
have you measured the current draw from the battery when it is running loudly? i'd be curious to know what that was.
the smaller cpu heatsink fans i have are rated at 60mA at 12V = 0.72W
having it on a thermal switch as you suggest would mean it only came on when things got hot enough to need it.
1.3 watts is 1/3 of what the amp uses at max power? that doesnt sound quite right... you must be using more than 3.9W! the amp6 basic is 25W per channel for a total max power of 50W. Tim is using an Amp9, for up to 4x100W per channel - i know he probably wont use a fraction of that power, but even so i think some cooling arrangement is wise.
have you measured the current draw from the battery when it is running loudly? i'd be curious to know what that was.
the smaller cpu heatsink fans i have are rated at 60mA at 12V = 0.72W
having it on a thermal switch as you suggest would mean it only came on when things got hot enough to need it.
Col, I’ve never thought about reusing old PC heat sinks before (there’s about three old PC boxes sitting in the garage at the moment) for this project or not it’s a good idea to have in mind. Thanks for the idea and the offer of the spare.
Saturnus, I can see what you mean, there will be very little room for air to move in the middle electronics compartment. Keeping the operating temp as low as possible will be very important (especially at festivals like falls, we had 40-45 deg C for three days last year). Thanks for the drawing you attached. I have a few comments/questions:
1. I can see what you did with the grooves in the middle. Very smart idea. The middle section is a lot slimmer than I would have assumed from the photos in the first post.
2. It looks like the woofer and breather hole cut outs are tapered, are they or is it just the view of the model? Is this necessary or purely aesthetic?
3. Two things are still unclear, the ports and the bracing in the bass reflex enclosures. Would I be able to get a copy of one of the drawings or CAD files that did survive?
By the way if anybody is interested, I have found the Aluminium enclosures here:
http://www.jaycar.com.au/productVie...&pageNumber=&priceMin=&priceMax=&SUBCATID=481
And the batteries here:
http://www.jaycar.com.au/productVie...&pageNumber=&priceMin=&priceMax=&SUBCATID=250
Saturnus, I can see what you mean, there will be very little room for air to move in the middle electronics compartment. Keeping the operating temp as low as possible will be very important (especially at festivals like falls, we had 40-45 deg C for three days last year). Thanks for the drawing you attached. I have a few comments/questions:
1. I can see what you did with the grooves in the middle. Very smart idea. The middle section is a lot slimmer than I would have assumed from the photos in the first post.
2. It looks like the woofer and breather hole cut outs are tapered, are they or is it just the view of the model? Is this necessary or purely aesthetic?
3. Two things are still unclear, the ports and the bracing in the bass reflex enclosures. Would I be able to get a copy of one of the drawings or CAD files that did survive?
By the way if anybody is interested, I have found the Aluminium enclosures here:
http://www.jaycar.com.au/productVie...&pageNumber=&priceMin=&priceMax=&SUBCATID=481
And the batteries here:
http://www.jaycar.com.au/productVie...&pageNumber=&priceMin=&priceMax=&SUBCATID=250
I can't remember the exact size actually. I know it's between 22 and 24 liters. And I can remember that I worked with either 22.65 liters or 23.45 liters but I can't remember which it was settled on.
Port size is 3407mm2, defined as 120mm total width and 30mm total height rounded to a 30mm circle in both ends. Length is 35mm, or 3 layered 12mm birchply which actually actual size is 11.65mm.
Here is the CAD files. It's a Google SketchUp file so you need to download that first. Please note that once it was imported there was some small changes for practical reasons. Such as making the port mounting slightly larger, not the actual port size but the wood pieces that make up the port.
The cutout in the top was removed and replaced with another cut out in one of the speaker enclosure size. This is because I first thought to stack 3 batteries which needed that cutout but in went with 2 batteries as it gives long enough battery time and it was also before I found that perfect amp6basic enclosure. The cutout in the top in one of the speaker enclosures is for the small electronics box that is on the backside of the solar panels.
I also changed the "breather"/access holes on either side of the middle section from being tapered to being rounded (3mm rounding) instead. The tapering on the woofers is essential in my mind as they are mounted from the inside but a heavy rounding (9mm rounding just as the tapering) could be done instead.
http://www.youshare.com/Guest/39078183761ead6c.zip.html
If you don't access to CNC machinery, I strongly suggest you look at making some changes that will enable you to make it using handtools. Such as reducing the number of interlocking routings which would be very hard to make accurately by hand.
Port size is 3407mm2, defined as 120mm total width and 30mm total height rounded to a 30mm circle in both ends. Length is 35mm, or 3 layered 12mm birchply which actually actual size is 11.65mm.
Here is the CAD files. It's a Google SketchUp file so you need to download that first. Please note that once it was imported there was some small changes for practical reasons. Such as making the port mounting slightly larger, not the actual port size but the wood pieces that make up the port.
The cutout in the top was removed and replaced with another cut out in one of the speaker enclosure size. This is because I first thought to stack 3 batteries which needed that cutout but in went with 2 batteries as it gives long enough battery time and it was also before I found that perfect amp6basic enclosure. The cutout in the top in one of the speaker enclosures is for the small electronics box that is on the backside of the solar panels.
I also changed the "breather"/access holes on either side of the middle section from being tapered to being rounded (3mm rounding) instead. The tapering on the woofers is essential in my mind as they are mounted from the inside but a heavy rounding (9mm rounding just as the tapering) could be done instead.
http://www.youshare.com/Guest/39078183761ead6c.zip.html
If you don't access to CNC machinery, I strongly suggest you look at making some changes that will enable you to make it using handtools. Such as reducing the number of interlocking routings which would be very hard to make accurately by hand.
About the power consumption.
The amp6basic has a power diode built in to protect it against reverse voltage so that drops the nominal voltage from 12 to 11.4 volts.
So the total maximum output power before clipping for both channels combined is (11.4^2/4*0.6) 19.5 watts on a nominal 12 volts. But that only what the peak output is because recorded music is a series of peaks and lows the average power comsumption is much lower. In fact most recorded music follow the RIAA standard dictating that the average sound level is mixed to 12dB below maximum peak level (0dB). And since we know that -12dB equal dividing with 4, the average output power before clipping is (19.5/4) 4.875 watts.
The effeciency of the amp6 is 81% in 4 Ohms so total power consumption at maximum power before clipping is (4.875/0.81) 6 watts precisely theoretically. Under real life condition I have almost never measured above 350 mA drain on a 12 volt supply equalling 4.2 watts when measured over several hours. That's probably because of silence between numbers and intro/outro fades etc.
Please note that the voltage will vary with the charge level of the batteries and so will the output power and consequently the power consumption as well.
The amp6basic has a power diode built in to protect it against reverse voltage so that drops the nominal voltage from 12 to 11.4 volts.
So the total maximum output power before clipping for both channels combined is (11.4^2/4*0.6) 19.5 watts on a nominal 12 volts. But that only what the peak output is because recorded music is a series of peaks and lows the average power comsumption is much lower. In fact most recorded music follow the RIAA standard dictating that the average sound level is mixed to 12dB below maximum peak level (0dB). And since we know that -12dB equal dividing with 4, the average output power before clipping is (19.5/4) 4.875 watts.
The effeciency of the amp6 is 81% in 4 Ohms so total power consumption at maximum power before clipping is (4.875/0.81) 6 watts precisely theoretically. Under real life condition I have almost never measured above 350 mA drain on a 12 volt supply equalling 4.2 watts when measured over several hours. That's probably because of silence between numbers and intro/outro fades etc.
Please note that the voltage will vary with the charge level of the batteries and so will the output power and consequently the power consumption as well.
Boominator box response
just stuck the TS parms in WinISD and run the numbers:
http://pix.minirig.org.au/main.php?g2_view=core.DownloadItem&g2_itemId=820&g2_serialNumber=1
col.
just stuck the TS parms in WinISD and run the numbers:
http://pix.minirig.org.au/main.php?g2_view=core.DownloadItem&g2_itemId=820&g2_serialNumber=1
col.
Sorry for slow reply, work has kept me too busy.
Thanks for the files Saturnus. Yes will definitely be eliminating some of the routed interlocking grooves, it would offer a very neat finish but is way out of my league with hand tools.
Cheers for modelling that col.
One question Saturnus, the components are all shown as individual panels and I am still not really clear how the bracing fits into the box?
Thanks for the files Saturnus. Yes will definitely be eliminating some of the routed interlocking grooves, it would offer a very neat finish but is way out of my league with hand tools.
Cheers for modelling that col.
One question Saturnus, the components are all shown as individual panels and I am still not really clear how the bracing fits into the box?
Are you sure? If you open the one that says BoominatorII.skp it should show all the panels assembled.
Anyways, yeah, sadly the bracing was the only part not to survive. It's pretty simple tough. It's 2 vertical panels of 12mm bircg ply running down the center following the grooves for them in the bottom, top, end, and center pieces.
Without CNC machinary to accurately cut it out from the start you should probably wait to cut out the holes in it until you have all pieces ready for assembly.
Or cut it into a large 0 shape 45-50mm wide on all sides except the port corner which is cut away.
Also get a sheet of 3-4 mm ply or masonite (or whatever you have handy), and use strips of that as guides. You'll need a piece for the bottom cover anyways.
Anyways, yeah, sadly the bracing was the only part not to survive. It's pretty simple tough. It's 2 vertical panels of 12mm bircg ply running down the center following the grooves for them in the bottom, top, end, and center pieces.
Without CNC machinary to accurately cut it out from the start you should probably wait to cut out the holes in it until you have all pieces ready for assembly.
Or cut it into a large 0 shape 45-50mm wide on all sides except the port corner which is cut away.
Also get a sheet of 3-4 mm ply or masonite (or whatever you have handy), and use strips of that as guides. You'll need a piece for the bottom cover anyways.
Saturnus said:
Btw, now I can remember it. The 22.65 liters was with the centre brace and the 23.45 was without it. That's because I made the first one without it but the testing showed that the top would vibrate too much under heavy loads if there was a full top solar panel on it. So the bracing was included on the next version. If you make it without the full top solar panel you don't need the bracing from a mechanical POW but I also think that version sounds better.
I calculated with 23 liters which is between the numbers for the with and without bracing. I figured that it wouldn't make a huge impact if it was slightly under- or over-tuned. And it don't.
Hi Everybody,
Hope you are all doing well. Once again sorry for slow replies and lack of progress updates, deadlines seem to pile up leading into Christmas. Amp6 and Amp9 both arrived on Tuesday. Had a chance to get to building them last night. I figured I would start with the Amp6 to “get my eye in” because it’s been a little while since I’ve done much soldering. Amp6 went together and was playing music in about 2 hours! Absolutely fantastic, very impressed with everything. Was missing one of the 3.3uF low ESR caps but I temporarily stole it from the Amp9 kit to get it going.
I then started on the Amp9. There is an awfully big gap in the level of detail in the assembly instructions between these two amps. The 41Hz forums are critical, especially the diode soldering thread (thanks col!) because they are seriously fiddly to get right. Amp 9 is up to testing stage. It was very late in the night by the time that was finished so I left it for another time, I would hate to do something stupid when connecting the power because I was too tired to concentrate. Should get to it tonight and fingers crossed all will be good.
As for the rest of the construction... I have decided to make the overall length 1m as per the rev 2 of your Boominator Saturnus. The model posted above appears to be the 900mm long version. This will give that little bit of extra space and hopefully room for more air to circulate around the Amp9B. The cut-out for access to the battery compartment will be on the top of the enclosure and the solar panel will be fixed in place with wing nuts. This will allow it to be removed for hot air to vent away from the amp if necessary and/or will include enough cable to place panel remotely if required (i.e. Boominator in shade and panel in sun).
A few quick questions about connecting the source to the Amp9B. I have a few options available; iPod classic (current gen.), iPhone 3G or iPhone/iPod to Headamp Pico. Most likely will be just be using the iPod as I would rather leave the Pico at home (one less thing to lose, work days would really drag without it) but can take it if there is the possibility of any significant performance gains. The two channels of the source output will be paralleled across two inputs each. So, for the questions:
1. Will the iPod Classic have enough “grunt” to drive all four inputs of the Amp9B?
2. How do I optimise the input sensitivity to best suit?
Thanks again everybody.
PS. Will hopefully get some progress pics posted over the weekend.
Hope you are all doing well. Once again sorry for slow replies and lack of progress updates, deadlines seem to pile up leading into Christmas. Amp6 and Amp9 both arrived on Tuesday. Had a chance to get to building them last night. I figured I would start with the Amp6 to “get my eye in” because it’s been a little while since I’ve done much soldering. Amp6 went together and was playing music in about 2 hours! Absolutely fantastic, very impressed with everything. Was missing one of the 3.3uF low ESR caps but I temporarily stole it from the Amp9 kit to get it going.
I then started on the Amp9. There is an awfully big gap in the level of detail in the assembly instructions between these two amps. The 41Hz forums are critical, especially the diode soldering thread (thanks col!) because they are seriously fiddly to get right. Amp 9 is up to testing stage. It was very late in the night by the time that was finished so I left it for another time, I would hate to do something stupid when connecting the power because I was too tired to concentrate. Should get to it tonight and fingers crossed all will be good.
As for the rest of the construction... I have decided to make the overall length 1m as per the rev 2 of your Boominator Saturnus. The model posted above appears to be the 900mm long version. This will give that little bit of extra space and hopefully room for more air to circulate around the Amp9B. The cut-out for access to the battery compartment will be on the top of the enclosure and the solar panel will be fixed in place with wing nuts. This will allow it to be removed for hot air to vent away from the amp if necessary and/or will include enough cable to place panel remotely if required (i.e. Boominator in shade and panel in sun).
A few quick questions about connecting the source to the Amp9B. I have a few options available; iPod classic (current gen.), iPhone 3G or iPhone/iPod to Headamp Pico. Most likely will be just be using the iPod as I would rather leave the Pico at home (one less thing to lose, work days would really drag without it) but can take it if there is the possibility of any significant performance gains. The two channels of the source output will be paralleled across two inputs each. So, for the questions:
1. Will the iPod Classic have enough “grunt” to drive all four inputs of the Amp9B?
2. How do I optimise the input sensitivity to best suit?
Thanks again everybody.
PS. Will hopefully get some progress pics posted over the weekend.
Hi col,
Yes, I was planning to get one when I head to Jaycar over the weekend. I'm still planning to try the different wiring configurations so even if I don't end up using it for this it looks like a handy (and fairly cheap device) to have around.
It does not look powered though so will it help with input power levels at all?
Cheers,
Tim
Yes, I was planning to get one when I head to Jaycar over the weekend. I'm still planning to try the different wiring configurations so even if I don't end up using it for this it looks like a handy (and fairly cheap device) to have around.
It does not look powered though so will it help with input power levels at all?
Cheers,
Tim
Hi Guys, I see some transformer pioneering around here and thought I'd let you know my last piece of pioneering with the Neutrik NTE-4.
If you have linear frequency response speakers and want to make the behaviour of the transformer linear too you should add a 470 or 680nF capacitor (ceramic quality is fine) and a resistor of 3,9 to 4,3Kohm in series and place this network in parallel with the secondary winding.
Some transformer datasheets show this type of correction, some not. Have a look:
http://www.farnell.com/datasheets/23903.pdf
(its at the bottom, but the correction they describe is NOT for the NTE-4)
I configured the network with a square wave generator and a scope and without the network you see terrible overshoots, with the network in place you get beautiful square waves!
If you have linear frequency response speakers and want to make the behaviour of the transformer linear too you should add a 470 or 680nF capacitor (ceramic quality is fine) and a resistor of 3,9 to 4,3Kohm in series and place this network in parallel with the secondary winding.
Some transformer datasheets show this type of correction, some not. Have a look:
http://www.farnell.com/datasheets/23903.pdf
(its at the bottom, but the correction they describe is NOT for the NTE-4)
I configured the network with a square wave generator and a scope and without the network you see terrible overshoots, with the network in place you get beautiful square waves!
Yes, all transformers give such behaviour...
I must add that the input impedance of the amplifier dampens most of it, but not all. It is however not do-able without the proper equipment to tweak it exactly right...
Here's what the measurements look like in general:
Bare wire:
Bare transformer:
With the correction it's pretty close to the original square wave again:
This provides a good starting point with nice flat response behaviour.
I can order a NTE1 myself and tweak a good working network and post my findings if you like Col?
I must add that the input impedance of the amplifier dampens most of it, but not all. It is however not do-able without the proper equipment to tweak it exactly right...
Here's what the measurements look like in general:
Bare wire:
An externally hosted image should be here but it was not working when we last tested it.
Bare transformer:
An externally hosted image should be here but it was not working when we last tested it.
With the correction it's pretty close to the original square wave again:
An externally hosted image should be here but it was not working when we last tested it.
This provides a good starting point with nice flat response behaviour.
I can order a NTE1 myself and tweak a good working network and post my findings if you like Col?