Go Back   Home > Forums > Amplifiers > Chip Amps

Chip Amps Amplifiers based on integrated circuits

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 21st January 2008, 09:51 PM   #1
slomatt is offline slomatt  United States
diyAudio Member
 
Join Date: May 2005
Location: San Francisco, CA
Default Toroid question and LM3886 photos

Greetings. I've been working on a LM3886 kit from chipamp.com and finally got around to putting things together. Brian's kit is nicely laid out and easy to build, it's been a fun project so far.

I will be using the amp to drive a set of 4 ohm speakers, so I'm trying to keep Vcc near 28v as per the national semiconductor data sheets. To meet this goal I'm using the XFM-D0012 toroid from apex jr which has dual 21.5v secondaries. Surprisingly, when I hooked it up I am getting ~27v AC per secondary, which leads to a rectified +/- 35v output from the power supply.

Is something up here? I can think of three possibilities:
1. I hooked up the toroid wrong.
2. My calculations were wrong. I was assuming 21.5v x 1.4 = ~30V rails.
3. The toroid is putting out a higher voltage than it should (27v vs 21.5v).

There were no instructions for the toroid, so I'm hoping someone can review my wiring scheme. Here's a picture of the toroid.

Click the image to open in full size.

Here's how I have it hooked up and also an explanation of my assumptions. This is for 120v mains. I'm running dual-mono power supplies with a separate toroid and ps board per amp.

I measured ~12 ohms between orange/red and also between brown/black, my assumption is that each pair is one primary. For 120v I need them in parallel so I hooked up orange to brown and red to black this results in an input resistance of ~6ohms which seemed reasonable. I'm also assuming that the colors are listed in order of winding so that orange->red is in phase with brown->black.

With the above wiring I measure 27v on each of the outputs (blue/black and red/white). I've wired blue to AC1 (line above), black to AC1 (no line), red to AC2 (line
above), and white to AC2 (no line).

Here is a schematic of the ps circuit:

Click the image to open in full size.

Can anybody verify that the above wiring is correct?

When powered up the blue LEDs light up and the amp seems to work properly, my main concern is the high rail voltages since I will be using a 4 ohm load.

During lower power usage the heat sink did not get above 100 degrees F (ambient temp around 65). I was using my laptop CD drive as a test source, and at one point the computer went to sleep and when I woke it up the lightbulb in series with the amp glowed brightly, there was a strange humming in the speaker, and the chip temperature spiked from 90 deg F to 100 deg F. Any idea what could cause this?

Any input or suggestions is greatly appreciated. I've been building speakers for years, but this is my first amp project and it's been quite fun.

Here are some pictures:

more here

Click the image to open in full size.

Click the image to open in full size.

Click the image to open in full size.

- Matt
  Reply With Quote
Old 21st January 2008, 10:28 PM   #2
gni is offline gni  United States
diyAudio Member
 
gni's Avatar
 
Join Date: Sep 2006
Two options:

(1) Very efficient circuit with little or no loss


(2) Mislabeled Transformer

I like number 2
  Reply With Quote
Old 21st January 2008, 10:33 PM   #3
diyAudio Member
 
danielwritesbac's Avatar
 
Join Date: Nov 2007
Well, rails of 35+35 DC should be perfect. . . the result is fine.

What's the AC voltage in your house?

That speaker on your desk, is it a b20?
__________________
Tools, Models & Software for DIYClipNipper boostLM1875 TurboPowerful TDA7293 kitTDA7294 pt2pt ♦ My post has opinion.
  Reply With Quote
Old 21st January 2008, 10:38 PM   #4
diyAudio Member
 
Join Date: Sep 2005
Location: Melbourne Florida
Default TWO HIGH

Your AC lines can very greatly from place to place and time to time. Ive had AC mains drop as low as 102VAC and go as high as 127VAC. Check your mains, if your trans primary is 110VAC and your mains are high like 120+VAC there is your difference.
Thats why most manufactures allot for +/-20% mains variance.
Your chip will work with a 4Ohm load at +/-30DC just make sure you have adequate heat sinks.
  Reply With Quote
Old 21st January 2008, 11:09 PM   #5
BWRX is offline BWRX  United States
diyAudio Moderator Emeritus
 
Join Date: Jan 2005
Location: Pennsylvania
Default Re: TWO HIGH

Quote:
Originally posted by tiltedhalo
Check your mains, if your trans primary is 110VAC and your mains are high like 120+VAC there is your difference.
The transformer label shows a 110/220V primary winding, so tiltedhalo's comment is right on the money. The voltage will also be a bit higher because the transformer is not being loaded down.
__________________
Brian
  Reply With Quote
Old 21st January 2008, 11:09 PM   #6
diyAudio Member
 
Join Date: Apr 2007
Another vote for: Check your actual AC outlet voltage. Mine is 126VAC+, yielding nearly 31VDC from a 20V toroid after all is said and done.


7/10
  Reply With Quote
Old 21st January 2008, 11:40 PM   #7
slomatt is offline slomatt  United States
diyAudio Member
 
Join Date: May 2005
Location: San Francisco, CA
Thank you for all the helpful replies, that's why this place is such a great forum

I'll measure the AC at the house tonight to see how close it is to 120v. I'll also measure the dc voltages under load to see what they are. The good news is that I only have ~100mV off offset without the feedback capacitor and just a little hissing noise from the speakers. Everything is good so far. I'm looking forward to finishing the second amp so I can test everything in stereo.

The "big" problem is that I'm not sure I'm going to be able to fit everything into the enclosures I want to use. My goal is to build two mono-blocks, but I may wind up with two ps units and a single amp unit, or possibly two ps units and two amp units. It seems a bit silly to go with a four enclosure fully separated solution for a lower cost amp like this. The enclosures are going to be the Ikea plate warmers shown in the picture with some metal grid below the top holes and solid maple sides.


danielwritesbac,

The speaker on the desk is indeed a B20, there's a more "up close" photo in the additional photos I linked to. Nelson Pass was very generously giving them away at Burning Amp. At some point I'm going to put them in a set of Curvy Chang enclosures.


- Matt
  Reply With Quote
Old 22nd January 2008, 01:14 AM   #8
diyAudio Member
 
danielwritesbac's Avatar
 
Join Date: Nov 2007
Did the label on that transformer really say 2.7 amps?
If so, that's for monoblocks, but sure you can put the transformer in the enclosure--just don't run the signal cables across it.

Ah, be sure to duplicate test bench conditions within the enclosure. This is regarding the ground lines. There's not just one. The speaker ground goes nowhere but the speaker. . . the input ground goes nowhere but the input. . . the CG common ground goes to the enclosure.

So, how does it sound? OH! I mean, how does it sound on the loudspeaker? We don't really want to make the B20 driver go hopping around all over the desk.

EDIT: That transformer can run stereo lm3886 if using the configuration suggested by chipamp.com.
Bit of Artic Ceramique thermal paste to heatsink it and it'll do fine.

EDIT: The grounding is something to watch on stereo config. Each individual speaker ground goes to its own speaker terminal.
Each individual signal ground goes only to the input--those wires are exactly the same length for the stereo pair.
The power lines are exactly the same length for the stereo pair.
The CG's, common grounds are exactly the same length for the stereo pair.
This and some luck will avoid a ground loop.
__________________
Tools, Models & Software for DIYClipNipper boostLM1875 TurboPowerful TDA7293 kitTDA7294 pt2pt ♦ My post has opinion.
  Reply With Quote
Old 22nd January 2008, 07:51 AM   #9
slomatt is offline slomatt  United States
diyAudio Member
 
Join Date: May 2005
Location: San Francisco, CA
The amps are wired and playing music (Dire Straits) as I write this. Here's a photo of the test setup:
Click the image to open in full size.

Source: Marantz DR 6000
Preamp: Bottlehead Foreplay w/ Anticipation Upgrade
Speakers: Adire Audio GrandPop! (4ohm)

Measurements: (channel A/channel B)
Wall voltage: 121.5v
Secondaries (no load): 26.9v/26.9v
Secondaries (load): 26.6v/26.6v
Rails (no load): 36.8v/36.8v
Rails (load) 35.3v/35.4v
DC Offset: 106mv/105mv

The chips are on undersized heatsinks and not yet mounted using any compound, the heatsinks are running ~101 deg F while playing at a low volume.

I was hoping for rails around 30v and wound up with 35v rails, so I'm a little concerned about issues with a 4 ohm load. Hopefully decent heatsinks will solve that issue. Is there anything I should watch out for as an indicate of amplifier distress?

The transformers put out 2.7A. Based on the equations at National Semiconductor I estimated the power requirements for 68w@4ohm to be 23.3v @ 5.8A and for 38w@8ohm they would be 24.7v @ 3.1A. I figured dual 2.7A transformers should be enough.

With no input connected I hear a slight hiss in each speaker from a distance of ~3". There is no hum, which surprises me since the amps are currently not in an enclosure and sitting on a cardboard box. Grounding is all handled through the board connections, I'll use a similar scheme in the enclosure but will of course make sure the chassis is safety grounded.

I did notice a slight AM radio pickup on the left channel, which I attribute to the complete lack of shielding of the circuit boards and the input lines.

My preamp has a 60Hz hum that I am in the process of tracking down. With my normal amp (amp section of a NAD 314) this is annoying but not too big of a deal. The chipamp has considerably more gain and the hum is very noticeable, I'll have to work on debugging the preamp soon. Another issue is that the tube preamp puts out a lot of voltage on the outputs, when coupled with the high gain of the LM3886 this means the usable range of the volume pots is about 1/4 turn. I might consider padding the input of the chipamp to compensate.

Ok, so enough technical business, how do they sound? (caveats: no enclosure, wired through ancient and corroded rca jacks, still using a 60w lightbulb as a fuse)

So far I am impressed. I'm not sure if I'm ready to say they sound much better than the NAD, but I haven't had a chance to turn up the volume yet. Imaging and "space" around instruments are very impressive. The sound is very clean, though I'm worried that the treble might be a little bit too thin since cymbals don't have a lot of ring or air to the sound. I'll have a much more detailed review tomorrow when I can turn up the volume a bit.

Thanks again for everyone's input and advice.

- Matt
  Reply With Quote
Old 22nd January 2008, 12:31 PM   #10
diyAudio Member
 
Join Date: Sep 2005
Location: Melbourne Florida
Default SOUNDS GOOD?

Just make sure your heat sinks are large enough for the load and voltage, take a look at nationals figures for dissipation and heat sinking.
I have tested the 3886 and 3875, I prefer the 3875 at +/-75VDC at low outputs with success, the little buggers get red hot but did not go into spike protection. Remember these chip are totally built on a very small die, thermal management is nationals biggest concern with supply voltage hence the low max supply not the transistors ability to handle it, but keep the dissipation down to a level that can be shed.
As for the sound, give it some time, they do need to burn in a bit. The treble will mellow out with time, if not try adding a 10uf silmic bypassed with a 1, .1 uf film and foil cap at the input.
  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


Similar Threads
Thread Thread Starter Forum Replies Last Post
Please Help Toroid Question for F1 milezone Power Supplies 6 27th August 2009 09:40 PM
My monoblock Gainclone LM3886 (photos link) steeve Chip Amps 7 11th March 2005 08:37 PM
New Pas Question no photos lpd Tubes / Valves 5 27th November 2004 07:06 AM


New To Site? Need Help?

All times are GMT. The time now is 03:50 AM.


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