diyAudio (
-   Chip Amps (
-   -   new amplifier for 1970s powered advents (

thetube0a3 7th September 2008 06:08 PM

new amplifier for 1970s powered advents
hey this is my first post so im sorry about its crudness :)

ive been repairing 1970s powered advents for awhile now and recently one of mine broke again. so i decided to try and build some chipamps for them.

with the help of Randy M Rubin "" i built these amplifiers.

i used a 500VA 25+25 toroid with a 50A fairchild bridge and two mallory 24,000uf 50v caps. and lets not forget the relay rated for a 25A inductive load :)

the board is 1.6mm with 2oz copper and black solder mask.
the tweeter amplifier is just a single LM3886 configured for 50W output into an 8 ohm load. the woofer amp is an almost exact copy of the BPA-200 servos and all.

a quick test showed it was flat from 20Hz to 100KHz, i didnt try below 20Hz.

i have 1 volt of ripple on each rail at full output, and 30-60mv of dc offset on the tweeter amplifier depending on temp\load. and between .5mv and 1mv on the woofer amp at all times

these bolt down to the orignal back plate of the powered advent, and will have 1"x1" aluminum bars milled smooth and bolted down to bolt the LM3886's too to sink them to the backpanel which is 90 degrees to them. sorry they arnt installed in the pictures, i had small 1"x1" heatsinks for quick testing and tweaking.
the new crossovers are 1.6mm 2oz copper red boards, that bolt down on standoffs near the center of the board, they are a 12db \ octave linkwitz riley designed to match the natural rolloff of the drivers, i also have a second 24db / octave one incase i want to use the same board on another speaker. perfect for biamping anyting ive got :) the amplifiers were also made so i can use them without the crossovers all together. let me know what you think

christopher jensen

thetube0a3 7th September 2008 11:26 PM

2 Attachment(s)
sorry, didnt add any pictures

thetube0a3 7th September 2008 11:26 PM

2 Attachment(s)
and the bottom

lineup 7th September 2008 11:46 PM

i have 1 volt of ripple on each rail at full output

how much current/total power drawn from supply at 'full output' ?

because 1 Volt RMS 120 Hertz ripple
can be made lower .. if you want to .. and think that you need it

1. just add two more rather big caps (say 2200/4700uF) at the LM3886 boards
that is, rather close to chip(s)

2. eventually put in 0.1 Ohm filter resistors (power resistors!)
.. in the wire in series right after your big blue caps

Otherwise :)
Congratuations to a great Audio Sound System.
And a good posted description, too.

thanks & regards

thetube0a3 8th September 2008 03:58 AM

theres a 1000uf 50v panasonic FM cap on each rail right by the chip as well as .1uf caps

the 1volt ripple per rail was measured on the big caps, not on the actaul board.
ill let you know on any updates, thanks for the resistor tip tho :)

lineup 8th September 2008 04:18 AM

You already have those 1000 uF. Good!
And of course, what counts is the RIPPLE at CHIP supply pins.
What CHIP 'can see' is only what is right next to the PINS.

Now, the resistor.

Why not only use:
CAP24000uF - WIRE - CAP1000uF
Why use:
CAP24000uF - 0.1 Ohm + WIRE - CAP1000uF

Because the wire alone has not very high resistance+inductance.
Especially if wire is THICK, like Loudspeaker cable for example.
Not high impedance compared to one RESISTOR 0.1 Ohm ( 5w/10watt power resistor ).

To make an effective FILTER, we can not use Cap alone.
Cap together with some resistance makes one FILTER.
So called RC filter.

In this case the two caps + some resistance between form one
C - R - C filtering of your supply.
These are very effective filters, but not if the resistance component is too low.

By adding some little resistance, we will reduce ripple a bit.
It would be interesting if you measure Ripple BEFORE and AFTER adding resistor.

So you can see if I was really right ;) ... or not.


thetube0a3 9th September 2008 04:11 PM

you know, the main thing i wish i could improve but dont know how is the dc offset on the small amplifier. its just a single lm3886 with a standard non-inverting circuit, 21 gain using a 20k and a 1k resistor in feedback, and i have a 20k pot on the input. and of course a 1k resistor in series with the + input and a large coupling capacitor at the very begging. if i turn the pot all the way down grounding the input of the lm3886, i have an offset of 16mv. if i turn it all the way up, leaving only 20K resistance to ground on the input, i end up with between 40 and 60mv depending on temperture and such. is there any easy way to improve this?

im only concerned because i plan on this directly driving my tweeters. and its alittle on the highside

thetube0a3 10th September 2008 03:33 AM

2 Attachment(s)
i just took this picture to show progress. the heatsink brackets? or whatever you want to call them will be milled and drilled tommorow. and i started working on the crossoverboard finally :)

still waiting on parts tho, wife wont let me have any money! :)

lineup 17th September 2008 09:33 PM

wife wont let you have money
to buy necessary things

You have to talk some sense into that woman!

Does she want one happy husband ... or a sad one
By the answer to this question, you will know
if she really still loves you
or not ... :(

(show her this post .. and we see what happens)

Lineup - happily un-married & sharing life with his amplifiers
.. instead

thetube0a3 17th September 2008 10:06 PM

lineup : now that you saw what the amp actaully looks like what do you think of it? its my first amp.

All times are GMT. The time now is 09:04 PM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio