Leach Super Amp Pcb Re-Design (LSAPRD)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
peranders said:
The wiring! => Eliminate the wiring!

PerAnders,

please, please, can you repeat that just one more time !
:clown:

I thought one board designs not only look better but are better because of the statement above.
Then there is Mikeks on the pcb design rule thread just posting that IHO it is better to separate gain and current stage pcb's completely.

However, some here prefer building a Super Leach with original TO3's, others with Plastic TO247/264, even one that prefers doing both and use difficult cases.
So, what is the answer that pleases the individual.

We can put the thumbscrews on Jens to do single board designs in all variety's, keeping him busy till next Christmas !

I think the answer to Terry's question is that there is something wrong with the original Leach boards :
- The Leach deserves a high quality board, a PerAnders quote.
- Jens's design incorporates everything the original Leach board does not have.
- The Leach design does not have output boards. If output boards are made in Delta Audio quality, it makes no sense using the original Leach board.
 
jacco vermeulen said:


PerAnders,

please, please, can you repeat that just one more time !
:clown:

I thought one board designs not only look better but are better because of the statement above.
Then there is Mikeks on the pcb design rule thread just posting that IHO it is better to separate gain and current stage pcb's completely.

However, some here prefer building a Super Leach with original TO3's, others with Plastic TO247/264, even one that prefers doing both and use difficult cases.
So, what is the answer that pleases the individual.

We can put the thumbscrews on Jens to do single board designs in all variety's, keeping him busy till next Christmas !

I think the answer to Terry's question is that there is something wrong with the original Leach boards :
- The Leach deserves a high quality board, a PerAnders quote.
- Jens's design incorporates everything the original Leach board does not have.
- The Leach design does not have output boards. If output boards are made in Delta Audio quality, it makes no sense using the original Leach board.


OK, please don't think I'm trying to be difficult, because I'm not. I'm just trying to learn.

First, what is wrong with the original Leach boards?
If Jens' design "incorporates everything the original Leach board does not have", is it still a Leach amp?

I just finished populating my Superamp boards and have the T03's mounted on a Channel that is going to bolt to a large heatsink. I have spent many hours contemplating how best to wire the two together. The truth is I wish it was all together on one board like my P101 was. That was so much easier.

This brings me to my next question.
I have seen talk about wire lengths. Just how long can the wires be between the driver board and the T03's before you get problems? Do they need to be twisted or does that matter for these?

This thread came along a little too late to save me this work, but it's all in the name of learning right now so I don't mind so much. I am very interested in hearing the differences between the ones with the flatpacks and the T03s.

Blessings, Terry
 
Wiring?
 

Attachments

  • steg1.jpg
    steg1.jpg
    65.2 KB · Views: 1,049
If the wiring were really such a big deal sonically, the builders of these amps probably would have noticed.
Other than the work involved and attention to detail, seems this is not really a problem.
Thus far not one accurate statement has been made about anything that is wrong with the original boards. There are folks who don't like them.
"I think the one board solution is better, since the long wires are eliminated. It does however need a bigger heatsink."
When you use plastic transistors, a larger heatsink is always required.
Personal preference or bias.
All the parts required to build this amp are readily available as you know from:
http://www.onsemi.com/ ............even free samples.
The original circuit board is 4" x 6".

Prosit
 
just talking about the boards, not the Leach circuit :

The original Leach board layout :

- 1-layer design, which makes signal travel longer.
- little to no ground tracing, with the risk of signal influence.
- ground plane rather small, risking ground potential and currents.
- with exception of the ground plane and two powerlines practically constant trace width everywhere in the circuit.
Meaning many traces are wider than should be, many too narrow.
- many 90 degrees corners, even electricity does not enjoy steep curves.
- some long trace lengths, keeping signal path short is the no1 rule. Much easier with dual layer pcb's.
- kind of sloppy made, a great many irregularities, some pads are too big, some too small. Why spend so much money on components and use a second rate quality board ?

Jens's layout of the extended Leach offered additionally:

- through holed, which guarantees top solder connections.
- solid 1/10" board.
- 3 ounce thick copper surface, try find such a board.

Everyone knows any amplifier design that uses hard wiring can be made to oscillate.
For over 20 years hard wiring is avoided with solid state at all cost, all manufacturers that did hardwire their amplifiers stepped up to full pcb use ages ago.
 
Regarding wires, to be honest I’m more worried about the fact that wires and their mounting are a source of error. The simplicity of having a single board appeals to me because of the following reasons.

I see the following pros and cons.

Advantages:

1) Simpler assembly
2) No wires of unknown length in the feedback loop, makes a more stable construction
3) No wires, this means one less risk of making an error
4) The heat sink and PCB is a nice building block for an amp
5) Very compact design
6) When you debug, you know the connections are ok, so you can look for broken parts or bad soldering
7) It looks better to me

Disadvantages:

1) Needs bigger heat sink (But the flat packs require a larger one anyway)
2) More expensive because of the heat sink
3) It might be more difficult to put in a box, because it’s less flexible then the original design

What did I forget?

\Jens
 
"many 90 degrees corners, even electricity does not enjoy steep curves."

The foil can be found here:
http://users.ece.gatech.edu/~mleach/superamp/layout3.pdf

decide for yourself where the many 90 degree corners are.

Advantages
1) Simpler assembly
2) No wires of unknown length in the feedback loop, makes a more stable construction
3) No wires, this means one less risk of making an error
4) The heat sink and PCB is a nice building block for an amp
5) Very compact design
6) When you debug, you know the connections are ok, so you can look for broken parts or bad soldering
7) It looks better to me

Issues having to do with attention to detail, preference or prejudice.
The other verbiage is only hubris.

Copper clad appropriate for this board is available many places in the world (even the USA) regardless of oz. weight.

What did I forget?
TO3 versions of the high voltage plastic transistor are also readily available.
Don't forget the religion that tubes sound better.

Prosit
 
A look at onsemi.com shows no 350V types in the TO-3 package.

Also the linearity of the TO-3 types looks to be a bit off compared to the new devises.

But since this threat has decided to use the flat pack types, maybe everybody that wants TO-3 packages should start their own redesign thread?

\Jens
 
Still, not one accurate statement has been made about anything that is wrong with the original boards.

MJL21194, MJW21193 are 250V

MJL1302A is 230V

TO3 versions of the high voltage plastic transistor are also readily available in this voltage.

truth \troo-th
Conformity to fact or actuality.
Synonyms: veracity, verity, verisimilitude
These nouns refer to the quality of being in accord with fact or reality. Truth is a comprehensive term that in all of its nuances implies accuracy and honesty, freedom from deceit or falseness

hubris \HYOO-bruhs\, noun:
Overbearing pride or presumption.
This is the actor's hubris, to imagine the world possessed of a single, avid eye fixed solely and always on him.
--John Banville, Eclipse

But since this threat has decided to use the flat pack types, maybe everybody that wants TO-3 packages should stop pointing out that the emperor is naked.

Prosit
 
JensRasmussen said:
A look at onsemi.com shows no 350V types in the TO-3 package.

Also the linearity of the TO-3 types looks to be a bit off compared to the new devises.

But since this thread has decided to use the flat pack types, maybe everybody that wants TO-3 packages should start their own redesign thread?

\Jens

I hate to say things twice, but for you I’ll repeat it.

The MJL4302 and 4281 are 350V types.

Try looking at the datasheets and compare them for some truth \troo-th :rolleyes:

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