Go Back   Home > Forums > Amplifiers > Class D

Class D Switching Power Amplifiers and Power D/A conversion

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 12th October 2012, 08:20 PM   #3281
diyAudio Member
 
danielwritesbac's Avatar
 
Join Date: Nov 2007
The TA2020 doesn't need a heatsink for 8 ohm loads, but it does need a heatink for 4 ohm loads. See the efficiency difference between 8 ohm versus less for 4 ohm in the datasheet. The difference must have a place to go, and that is a heatsink, else the chip burns up when driving a 4 ohm load, such as Boominator.
__________________
Tools, Models & Software for DIYClipNipper boostLM1875 TurboPowerful TDA7293 kitTDA7294 pt2pt ♦ My post has opinion.
  Reply With Quote
Old 17th October 2012, 11:02 PM   #3282
diyAudio Member
 
Join Date: Oct 2012
I've read quite a bit of this thread and I'm planning on building out a Qubinator over the next 6 weeks but I have a couple simple questions:

1. When using the TA2024 MKII with two full range Goldwood GW-1058's, do I simply wire each speaker to it's own channel or do I need to put them in series?

2. Since the depth of the Goldwood GW-1058's doesn't allow the magnets to touch, do you use the masonite as a conjoining piece? Also, do you epoxy the subs to the center piece?

3. I was planning on using this battery pack New DC 12V Portable 9800mAh Li ion Super Rechargeable Battery AKKU Pack for CCTV | eBay (2V Portable 9800mAh Li-ion). Does anyone know of a battery level gauge or LED that is compatible with this kind of pack?

Thanks guys. This post is great... and one of the longest I've seen.
  Reply With Quote
Old 18th October 2012, 09:34 AM   #3283
ron1 is offline ron1  Ireland
diyAudio Member
 
Join Date: Sep 2012
Quote:
Originally Posted by Saturnus View Post
You don't have to use a heat sink. But it's highly recommended. Or as most, use thermal double adhesive tape and tape it to the side of the cabinet you're using.
Would this heatsink be overkill?
ImageShack® - Online Photo and Video Hosting

I tested the system last night to make sure everything worked. It went fine but there wasnt much bass and the highs seemed very high! All the woofers were just siting flat on the table so im sure they will sound good in the enclosure.
ImageShack® - Online Photo and Video Hosting

I have 150 ohm carbon resistors on the tweeters so i might adjust them. I know i must play pink noise through the tweeters so im just going to download a pink noise audio file and play it through the amp 6 with only the tweeters connected.- Question.
Will i keep the 150 ohm resistors connected to the tweeters and is there a better way to play pink noise through the tweeters as i have lots of audio equipment at home. Thanks.
  Reply With Quote
Old 18th October 2012, 06:09 PM   #3284
diyAudio Member
 
Saturnus's Avatar
 
Join Date: Jun 2005
Location: Copenhagen, Denmark
Yeah. It's overkill but you can use it. I'd search my (friends) computer scrapyard for an old motherboard cooler. That'd be optimal.

You don't have to play pink noise through the tweeters. Just be aware that they wont sound properly matched with the woofers until they've played a couple a day or two then.
  Reply With Quote
Old 18th October 2012, 08:09 PM   #3285
diyAudio Member
 
danielwritesbac's Avatar
 
Join Date: Nov 2007
The heatsink can be scrap metal, flattened part of an old aluminum frying pan, and either hotsink (or heat spreader) or heatsink methods, as long as the result is 4x bigger than the chip, AND is for the purpose of moving the heat away from the chip.

Arctic Ceramique thermal paste (computer supply) is very helpful!

Let's see, two of 4 ohm speakers, ~85% efficiency (means 15% heater), and that's about 1cm x 3cm x2cm at bare minimum, and if the PCB also helps out.

The location of the heatsink, vertical atop the typical TA2020 PCB, without air intake vents provided through the PCB, means that the deal is best used sideways so that there's both air above and air below the heatsink (to allow for natural convection airflow--hot air wants to go up, but only if there's cooler air to replace it).

Gluing the PCB to the side of the cabinet looks pretty good.

However, for horizontal mounting (bottom or top of the cabinet, a "Driver" heatsink can work. Old school amplifiers had a separate driver heatsink that never worked as a straight up wall. What we did for an upgrade was to replace it with "T" channel or "L" channel, whereby the bottom leg attached to the active device and the top shelf got a lot of little holes through it to allow air to go through it. The position of barely elevated up and over the PCB did get some airflow, which was better than the heat pooling situation it had before.

On the topic of heat pooling breakage, I'm afraid that some of the prefab "boxed" enclosed amplifier products simply turn into a daft EZ-Bake Oven once taken outdoors in sunny weather. Those amplifiers could survive better naked. Alternatively, one could put 80mm holes and corresponding 80mm fan grilles into the aluminum cabinets (both top and bottom or at least on bottom), as well as a row of 1cm holes along both sides. Ventilating the amplifier cabinets could be done reasonably attractively. If one must use a cabinet, let it not be an oven.

The possibilities of amplifier failure are severe with Boominator, due to the assumptions about Class D heatsinking conflicting with the reduced efficiency when driving a 4 ohm load and outdoor summertime use (kerblooie!). See the datasheet. You're not having your cake and eating it too. It is barely better than some of the automotive Class B. In fact, my Philips car chip runs cooler than my TA2020. Just don't forget the heatsink and ventilation for that TA2020 if he's driving a pair of 4 ohm loads.

This situation is not as severe when driving an 8 ohm load of halfinator/qubinator. The 8 ohm load is much easier and cheaper--just gain mod and then glue a super loud Sure TA2024 (or more voltage tolerant and much easier to mod Muse TA2021) onto the side of the enclosure and allow the wide surface area to remove enough heat.


Since we do have a full bandwidth amplifier driving a part bandwidth speaker, we can drop the load on subharmonic to remove some waste and remove a little bit of waste heat. . .

Drop the load trick for Boominator:
Located at one of the woofers, a 1000u series capacitor (output cap) drops the load for pitches below 80hz. For a capacitive load like that, you do need it on the opposite end of the speaker cable--away from the amplifier. Feel free to fine tune the capacitor value for best bass.

Drop the load trick for Qubinator/Halfinator:
Located at one of the woofers, a 2200u series capacitor (output cap) drops the load for pitches below 80hz. For a capacitive load like that, you do need it on the opposite end of the speaker cable--away from the amplifier. Feel free to fine tune the capacitor value for best bass. This is especially effective on sealed box builds where the loss of electronic dampening on low pitches will cause some bass extension and thus a highly competitive Qubinator with cheap thunder added.

Dropping the load on subharmonic is something I propose for cleaner vocals and conserving battery power. As a cost, it will alter your low frequency tuning a bit. Not every track has subharmonic content, and there's no need of flattening the battery and straining the amplifier for content the speaker won't output. As a benefit, there's less cone breakup, and that means clearer vocals. Although highly controversial, I have personally found this (output cap, from amp output+ to speaker+) approach of great benefit for portable audio.

Conversely, trimming the input cap size down is somewhat similar, theoretically of greater benefit, miserable in practice, and does nothing helpful for standby/bias efficiency of portable audio--The only way to get the amp to see an easier load for boominator, is an output cap (or 16 ohm speakers is a effective theoretical alternative). Those woofers are in parallel. Cable to the farthest woofer and locate the output cap there (the amp wants to drive inductive load, and the speaker cable will make do).

Continue to have the tweeter cable attached to the woofer terminals. That tweeter is capacitive and will slightly help to damp the woofer cone movement (KUDOS to Saturnus!!). Unfortunately, with my added cap, the treble has now run through two capacitors (first the large cap that I've just now put series to the woofer and then next, possibly a tweeter cap). Not all large caps are capable of transporting hi-fi treble, and in this case, attempt to patch by paralleling an additional electrolytic in the range of 1u to 10u onto your 1000u to 2200u (makes a range of 1001u to 2210u series to the woofer).


Drop the workload on superharmonic too? At approximately volume potentiometer location or RCA jack location, you can put a 220pF capacitor from + to ground (one cap for left and one for right). Hopefully this will dampen the cables a little bit, so that the amplifier doesn't attempt to amplify junk that's higher than the audio band. Outdoors, the amp is exposed to a maximum amount of RF pollution and we'd probably like to try to ground some of that out, in order to decrease the amplifier's workload.


So, heatsink, heat transfer (away from the amp), and confine the amplifier's workload to only what the speaker is able to reproduce. This suite of problem reduction will make for longer battery life, less clipping, clearer vocals and a longer lasting amplifier. That makes loud playback more successful.


P.S.
Still working on simplified charge regulator. . .
__________________
Tools, Models & Software for DIYClipNipper boostLM1875 TurboPowerful TDA7293 kitTDA7294 pt2pt ♦ My post has opinion.

Last edited by danielwritesbac; 18th October 2012 at 08:23 PM.
  Reply With Quote
Old 18th October 2012, 08:28 PM   #3286
H3x0S is offline H3x0S  Croatia
diyAudio Member
 
Join Date: Oct 2012
Can't you get speakers similar to those much cheaper?
  Reply With Quote
Old 24th October 2012, 06:13 PM   #3287
diyAudio Member
 
Join Date: Jan 2012
Hi, about the bipole idea, as Saturnus mentioned earlier in the thread, one of the benefits is to achieve an "acoustic" bass boost due to that low frequencies are less directional than high frequencies. In the Boominator case I think Saturnus wrote that this was taken into account in the box design to keep the enclosure volume small instead of changing the frequency response from flat (beautiful idea IMO).

However, how large is this low end boost, and how does it depend on frequency? I guess there must be a fairly easy formula for this....?
I guess it is similar to quarter space loading compared to half-space for a monopole speaker, correct? Like 3 dB in the low end, but where does it roll off upwards in frequency?

I just ask out of curiosity.
  Reply With Quote
Old 24th October 2012, 09:05 PM   #3288
diyAudio Member
 
Saturnus's Avatar
 
Join Date: Jun 2005
Location: Copenhagen, Denmark
Quote:
Originally Posted by mrduke View Post
However, how large is this low end boost, and how does it depend on frequency? I guess there must be a fairly easy formula for this....?
I guess it is similar to quarter space loading compared to half-space for a monopole speaker, correct? Like 3 dB in the low end, but where does it roll off upwards in frequency?

I just ask out of curiosity.
The boost is actually a lot more in the low frequencies but that's because there are several factors that plays a part.

+6dB boost from placement directly on the ground. +3dB from bipole summation. And +3dB from tuned port at roughly 144hz.

Now not all boosts are in play at the same time to the full potential but an average of around +6dB can be observed over most of the frequency range from the bipole summation frequency and down to the f0 frequency.

Bipole summation frequency is speed of sound divided with distance between acoustic centers in meters.
  Reply With Quote
Old 25th October 2012, 06:50 PM   #3289
diyAudio Member
 
Join Date: Jan 2012
Quote:
Originally Posted by Saturnus View Post
Bipole summation frequency is speed of sound divided with distance between acoustic centers in meters.
Ah yes this refers to the phase, right? Half of an octave above bipole summation frequency there should be a dip I guess, but maybe small due to that such higher frequencies are more directional?

Anyway, thanks for the answer. And amazing thread and community around a single design!
  Reply With Quote
Old 26th October 2012, 02:11 AM   #3290
diyAudio Member
 
Saturnus's Avatar
 
Join Date: Jun 2005
Location: Copenhagen, Denmark
Quote:
Originally Posted by mrduke View Post
Ah yes this refers to the phase, right? Half of an octave above bipole summation frequency there should be a dip I guess, but maybe small due to that such higher frequencies are more directional?
Correct. The dip however is precisely compensated by the overshot on the bafflestep when the cabinet is a square on it's most dominant axis (since the distance between acoustic centers is fixed at half the length of that square given that the acoustic center on a bipole is always the exact center of the speaker on the most dominant axis).

Last edited by Saturnus; 26th October 2012 at 02:20 AM.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



New To Site? Need Help?

All times are GMT. The time now is 01:50 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2