|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
||Thread Tools||Search this Thread|
|29th January 2009, 01:24 PM||#1|
Join Date: Oct 2008
gainclone amp output issue
Greetings to all,
I am currently in the process of putting together my first (hopefully of many) gainclone.
At that, following various and numerous designs and implementations, I have encountered a particular issue twice with two chips: a full 24V is output from the speaker connection while no input is applied.
I will leave it at that and welcome and appreciate any suggestions.
Thanks in advance.
|29th January 2009, 02:45 PM||#2|
Join Date: Feb 2003
Location: Somerset, SW England
You will need to give us full details of your circuit, including the power supply!
The truth need not be veiled, for it veils itself from the eyes of the ignorant.
|30th January 2009, 03:33 AM||#3|
Join Date: Oct 2008
In regards to the 2? volt DC issue
Greetings to Nuuk and all others who might be reading tihs post,
In relation to details of my attempted gainclone:
After (the following) connections are made, a constant 24, 25 or even 26 volts DC comes out of the speaker out lead. (Voltage fluctuation, I take it, may be due to area demand.)
It begins with an EI core transformer being 230 volts primary with 18-0-18 (blue wire - black wire - blue wire) secondaries - center tapped and rated at 100 watts.
The secondary's blue wires are directly soldered onto a square type bridge rectifier's AC input. Rectifier is rated at 10amps for 100 volts.
With four 4,700uf capacitors rated at 50 volts, the bridge is connected to it using the following method:
The positive (+) pinout from the bridge goes to the positive (+) leads on two of the capacitors. These capacitors are now connected to a second pair of capacitors with the first pair's negative (-) leads connected to the second pair's positive (+) leads. The second pair's negative (-) leads loop back to the bridge's negative (-) output.
The center tap (black wire) from the transformer's secondary is lead - and directly soldered - onto the capacitor banks center row where the positives (+) and negatives (-) meet.
May I suggest it is a typical low-tech PSU setup.
Verified voltage entering the bridge end sits at above mentioned 18-0-18 volts.
Verified voltage exiting the capacitor end stands at 26.8 volts DC from the center tap to either side lead, and 53.6 volts DC accross the positive (+) to the negative (-) leadouts.
Polarity has been confirmed using a multi-meter.
Thence it is connected to the LM3886TF chip:
Please note, all connections to the LM3886 are done using 20cm leads directly soldered (onto it) using a low heat setting on the soldering tool. Wires are color coded with insulated alligator clips. The chip is mounted on a heat sink measuring 11cm by 11cm with 2cm stand-out fins.
Pinouts are wired as follows:
Pin 1 - red wire.
Pin 2 - NC.
Pin 3 - yellow wire.
Pin 4 - orange wire.
Pin 5 - red wire.
Pin 6 - NC.
Pin 7 - brown wire.
Pin 8 - blue/white wire.
Pin 9 - green wire.
Pin 10 - white wire.
Pin 11 - NC.
Connections as follows:
Red wires (pin 1 and pin five) are clipped onto the positive side of the PSU cpacitor bank. A 100uf capacitor's positive lead connects to this wire with the negative lead to the brown wire being pin 7.
Yellow wire (pin 3) goes to a parallel connection consisting of a 0.7uh inductor and a 10Ohm/3w resistor. A 8Ohm dummy load with a muiti-meter set to AC or DC volts represents a speaker. Speaker negative grounds to the brown wire being pin 7.
Orange wire (pin 4) goes to the negative side of the PSU capacitor bank.
A 100uf capacitor's negative lead connects to this wire with the positive lead to the brown wire being pin 7.
Brown wire (pin 7) clips onto the 0 volt lead from the PSU capacitor bank, it also serves as ground.
Blue/white wire (pin 8) connects to a 20k resistor which goes onto the orange wire (pin 4) and sits between the LM3886 and the 100ufcapacitor.
Green wire (pin 9) loops to the yellow wire (pin 3) with a 20k resistor between them. This lead is also grounded with a 1k resistor and a 10 uf capacitor.
White wire has no connection to anything. Not grounded. (could this be the source of the problem?)
I would like you all to pardon any redundancy and I hope my stated details did not entail confusing anyone.
I must say I tried variations of the above implementation four times on veroboard with a previous chip. The 2? volt DC issue was always present, and on the last attempt smoke and heat was (also) present.
Pin 10 was not ever connected (again, was/is that/this the source of the problem?)
The new connections are now on a second chip, of course. Both chips are of the TF topology which I take it to mean insulated packages.
Thanks for your assistance.
p.s. apart from the first chip's meltdown heat was never observed on the chips.
|30th January 2009, 06:59 AM||#4|
Join Date: May 2008
Possibility 1 is, you have unvoluntarily shorted pin 1, 4 or 5 to pin 3 and thus connected the speaker output directly to one of the rails.
Possibility 2 is, the IC is blown.
Possibility 3 is, you have DC at the input. Which is quite improbable, because you have not yet connected the input. Pin 10 is the input. Connect a 1 k resistor here. The other leg of the resistor should go to your potentiometer wiper or input connector, if you work with a preamp.
Ci with 10 µF seems very small. You should replace it later with a bigger value to get decent bass performance. For now leave it in, because it has nothing to do with the fault.
If you've always done it like that, then it's probably wrong. (Henry Ford)
|30th January 2009, 12:25 PM||#7|
Join Date: Oct 2008
Thanks so much for your suggestions.
Fortunately for me, the chip was not blown like the previous unit.
A 1k resistor between Pin 10 and ground did it.
My main speakers are 6 Ohms. I paralleled both (3 Ohms,) powered on the chipamp - after making sure the previous issue was no longer present - and was pleasantly surprised.
This little device is tremendously powerful running at 3 Ohms. It did not even get slightly warm after playing since the past two hours.
My next objective is to build six chip amps in order to tri-amp my main speakers. I'm thinking a single chip unit for the highs, a double chip unit for the midrange and a triple chip unit for the lows per side.
On a second note, what would be the highest value for capacitors "Cs".
|31st January 2009, 12:45 AM||#8|
Join Date: Oct 2008
I wish to let all know that I have had my test chipamp running for the past 15 hours with no issues.
As it is, the unit is connected to the left channel of a CD player only, it is after all a single chip unit for test purposes.
It is powered by an old transformer I found is a spare room, rated at 18 - 0 - 18 and 100watts.
The output of the amp is connected to both my main towers. Towers are rated at 6 Ohms, so paralleled, that makes it a 3 Ohm load.
In the begining, the highs was a bit prickly to the ear, however, after about an hour, it smoothed out quite well.
I am not able to run full volume, not because of limitation, but due to the massive power put out by this single chip.
I have not seen the circuit protection activated in any way.
The chip does not feel warm to the touch.
Remember, this was just a hasty put together for curiosity rather than anything else.
I consider this post closed.
Thank you all for your thoughts.
|Thread Tools||Search this Thread|
|Thread||Thread Starter||Forum||Replies||Last Post|
|Gainclone troulbe (LM1875 distorsion issue)||haft||Chip Amps||10||25th February 2007 08:29 PM|
|I have an odd transformer issue (want to build the gainclone)||Jerry Parker||Chip Amps||9||22nd October 2003 12:37 AM|
|gainclone sub amp/transformer issue||trespasser_guy||Chip Amps||1||28th May 2003 09:30 PM|
|Experts needed: Crazy phase issue with Inverted Gainclone||PaulHilgeman||Chip Amps||30||20th February 2003 06:06 AM|
|New To Site?||Need Help?|