diyAB Amp The "Honey Badger" on line grid board

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi,

I am about to build the Honey Badger on a line grid verobard. Why? Two reasons:
1) The PCBs are quite expensive
2) I am used to line grid, like the layout work and the idea of line grid

Well, the layout is finished. I will use double euro board (230mmx160mm) to be able to solder all transistors on the board. Its a stared GND layout and high current lines will be strengthen by copper wire the whole way.

It seems that this format is no more available in the world. Please correct me if I am wrong! Fortunately I have 3 boards left (about 30 years old).

The relay below in the layout is the turn-on-delay (one for each channel).

I you are interested I can post the next steps and my test result. But would be great if you share your thoughts.

Here the layout (i know, hard to read):
DiyAB Amp Honey Badger on double euro board layout - Album on Imgur
 
So, this weekend I could almost finish the first channel. Only 2 resistors and terminals are missing which I forgot to order. Gonna test ist by next weekend.

I would like to know your thought about line grid board. Even in earliest Elektor magazines (germany) a PCB layout was always part of the project. Seems that line grind is infamous, but why? Sure, some designs are critical to oscillations. I hope the DiyAB is rather undemanding. Will see if I encounter oscillations...

Here the picture:
http://imgur.com/a/x6eqd
An externally hosted image should be here but it was not working when we last tested it.
 
Last edited:
After commissioning was interrupted for a long time because my wife forced me to do some renovation work, I could now found the time to continue with the project.

After the corretion of a few minor mistakes, the amp is working now pretty well. The only thing is that is seems to be on the edge of instability. At the first power up the resistor which shortcuts the oscillations heatet up very much. But this appeared never agaim. Sporadically there are oscillations around 1V but the resistor never really heated up again. Nevertheless, I have ordered some 2.2pF capacitors as lead compensation, I hope this makes the amp more stable. If not I would try it with CMC in addition as proposed.

In the next step I will redesign the layout to fit on euroboard since the huge and very old boards are scrap and no more available.
 
Well congratulations.
Another iconoclast doing it his own way.
Use what you've got, buying new electronic everything every two years is what the Lords of Commerce have planned for our lives. ***** them.
I built an amp channel point to point, a technique that went out of fashion about the time the labor rate went over $1 per hour. I'm listening to it right now, Berlioz Carnival Overture.
I'm retired, what do I care about labor efficiency. Beats playing Parchisi at the senior center. (kiddie board game)
If you're vaguely interested, see pictures on post 212 of this thread:
http://www.diyaudio.com/forums/solid-state/236256-retro-amp-50w-single-supply-20.html
Last page still I think.
I'm building an IC 20W amp today to replace the sound system on my 8 year old television. Typical dried up coupler cap fuzz in the sound now. **** C****. I had to buy a new tuner last year, couldn't get around the planned obsolesence dead remote control on the old one, but I can do a sound system myself. 30 v power transformer was surplus 25 years ago. Heat sinks out of a 50" LED TV found on the curb last summer with the fuse wrapped in aluminum foil - another early casualty of the consumer **** we're supposed to buy.
Best of luck with your efforts.
 
Last edited:
Veroboard isn't the best way in terms of parasitic effects, indeed. But you can follow some design rules as well to avoid that as far as possible. Especially, I am taking care of keeping the signal pathes as short as possible by using breaks.

The current and thin copper issue I solved by using copper wire in addition for all high current lines. And this problem you also have on PCBs where the track width is also limited.

I am now working on a redesign for euroboard (160x100mm). But I have to split the circuit because all the power transistors won't fit on the smaller board. The picture shows where I want to seperate VAS from the power stage. Is this correct?

And since I am using shielded audio cable everywhere for the input signal, shouldn't I use it for the (high impedance) feedback path as well having separate boards for VAS and power stage?

I hope you can give some advice.

@indianajo, interesting technique I know only from tube amps, but I am 34 years old and generation transistor. Never build anything with tubes so far :)

An externally hosted image should be here but it was not working when we last tested it.
 
Looks like you have separated out the front end from the power end correctly.
I suggest you use a dual inline (D.I.L.) plug and socket to join the two PCBs. They come in straight and 90° versions @ 0.1" pin pitch.
Use multiple paralleled pins for each connection. Maybe use 4 pins for the Power Ground connection. This will give a lot of mechanical connection to resist vibration, etc.
R32 & 33 could be on either PCB. Maybe even place half values on both boards, to help avoid damaging shorting of a supply rail.
Add 1n400x power diodes to the front end board. Just like D8 & 9 to help avoid polarity issues when connecting. These diodes turn on the Bulb of the Mains Bulb Tester if you connect wrongly.
There was a discussion about parasitic capacitances to the long feedback link, (in another Thread), but we never reached any obvious conclusions.
 
Hi Andrew,

thanks for the feedback! But how about the shield for the feedback path? I am thinking of using shielded cable for the feedback signal and DIL for the rest. The shielded cable I can then directly route to where the signal is needed and avoid going across the whole board.
 
....................
There was a discussion about parasitic capacitances to the long feedback link, (in another Thread), but we never reached any obvious conclusions.

................ But how about the shield for the feedback path? I am thinking of using shielded cable for the feedback signal and DIL for the rest. The shielded cable I can then directly route to where the signal is needed and avoid going across the whole board.
I already addressed that.
You could look for the discussion, but I can't remember which Thread it was in.
 
I flew my feed back path in insulated wire over from the output of the speaker cap. About 6" long. No problems really, except the highest of tinkly bells are slightly distorted. Piano top octave on speakers my point to point board does very well, as well as the original dynaco ST120 circuit with the djoffe idle bias control board added.
I'm Buying faster VAS & driver transistors (than 3 mhz Ft) this week to see if the tinkly bells will straighten out and fly right.
I made the driver lines and heat sense circuit lines over to the output transistor heat sink in bare wire, too, about 5" long so I can turn the driver board over & work on it without unsoldering. No problems. I did put a 8 turn coil on the heat sense coming back from the output transistor assembly since it goes to a high gain base of the lower driver transistor.
 
Last edited:
I don't find loop area has anything to do with hum or distortion in the high current low gain areas. And emitter resistors are on emitter followers, about as low gain as you can get.
I was careful and twisted the input wires to the amp. The temperature sense wires enclose about 10 sqin of area. And the amp is dead quiet without input. Of course the power transformer is E type, enclosed in a steel case, and 8" away from the driver board, 13" away from the high gain input transistor, 12" away from the VAS transistor
 
On the well-known ELEKTOR Crescendo amp I got informed of stability issues, i.e. oscillations, suspected to be caused by exactly those ceramic 0.22R emitter (or source, btw.) resistors the OP has shown in his pics. Even ELEKTOR then has published some work arounds to cure this problem. So I recommended the OP to get rid of these WW resistors. That's all.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.