Yet another A-75

[jgeissin]

This project is a total leap of faith; but thanks to the designers, I am sure that it will operate first time and with little or no effort!
I am in the final construction phases of build two stereo A-75s, two for bi-amping purposes. I hope to earn the PASDIY label in the process. These are being made with the IRF540N/IRFP9540N TO-220 devices (I got 200 each so that I can ensure close tolerances after matching (I HOPE)). I designed the case originally with the power supply being separate, but soon decided on the all-in-one design.
The case is basicly a cube in shape, the heat sink being mounted inside the case, with a rear mounted fan (attempting to make it variable speed). The fan will push air in over the amp and PS PCBs, providing cooling for them, then be routed out both sides, over the heat sink fins, and out of each side at the top. The case will be sealed in the end to prevent leaks, which will eventually lead to dirt accumulation and reduce the efficiency of the cooling design (on paper anyway!!).
I am building the amp per all the articles that I have read, with liitle or no modifications.
Four filter caps (computer grade) are being employed for C10-12, 22K UF each. All other componets (with few exceptions) are 1% devices.
The output devices will be mounted to the heat sink, and will provide support for the PCB, that will also carry the output device's biasing resisitors. R31 and C8 will be mounted on the output terminals/binding posts.
Attached is a PIC of the chasis sitting on may saw table. It is made mostly of .125" (1/8") Aluminum plate/sheet (eBay). I discovered some time ago that aluminum can be worked similar to wood, only you must use carbide tiped blades/bits; not a problem! The front panel (not shown yet) was made through FrontPanel Designer (not a bad service!). I need to learn to paint aluminum, not as easy as it seems!
Internally, the four caps (C10-12) sit in each corner, the toroid in basicly the center, and using Aluminum dowel and some threading, I mounted a strip of Aluminum that runs down the center and in the front, forms a T for wire management. The main AMP PCB is being mouted in a fashion that it will hang in front of the fan. The PS PCB is mounted on the toroid's mounting plate (pop riveted to the plate are two Al channels). The internal fuses are mounted on the T rail. The output devices mounting holes are drilled, but not yet tapped (waiting for a bottom tap to get to the house!).
A star ground will be employed behind the toroid so that all grounds return to the same place. Thermal coutouts will be used to keep things from destroying themselves and a thermal switch to turn on the fan (no idea of how the thermal management is going to work in reality, need to get it built and test to see where it will go). A pot will be inserted into the fan circuit (the base of Q1 in the power supply circuit) to adjust the fan speed to only what is required. I want to cut the fan noise to a minimum.
I am building the test circuit that Mr. Pass has in one of his papers to test the 400 transistors for the output. That will take some time!
More pics can be seen at my website, www.geissin.com go to Photo Gallery, Jon, Jon's audio projects, A75 Class-A amp project.
More as it comes together, I hope to start actual building this week, starting with the power supply. Then the main amp PCB.
My thanks to Nelson Pass and Norm Thangard for the design and support of the DIY community!

[Nelson Pass]

[jupiterjune]

Well, I have a few words of wisdom to offer..

there is a line in the A-75 article, regarding setting the bias, that says something to the effect of ' once one side is operating correctly, disconnect everything and follow the same procedure for the other channel'

well, don't try any shortcuts there!

Well, my sump pump is having a rough day, I'm off to siphon all the water out of my back yard.

Enjoy the A-75--it is an awesome amp.

JJ

[jupiterjune]

btw

There are some mistakes on the AX boards

JJ

[BobEllis]

Originally posted by jupiterjune
btw

There are some mistakes on the AX boards

JJ

You mean A75 boards, right? Yes, there are a couple boo-boos on the audioxpress boards.

Trace the connections on the main boards to verify that the part you are soldering in connects to what you think it does. IIRC, there are two spots labeled R26 .

The PSU board error is in the fan circuit. The pass transistor is turned on hard as labeled. Full rail voltage on your fan makes it go fast There is an unlabeled hole on the board to correct the connection.

I also built a pair of A75s for biamping. Prepare for a WARM listening room full of wonderful sound.

[jacco vermeulen]

Considering just how many commercial and diy designs have been copycatted for profit reasons, i'm still frequently surprised not having heard of anyone jumping the gun to serial manufacturing the A-75 design.

I've just spent a fist full of dollars on exotic obsolete Japanese semis to pimp JAM Sumasundram's A-100 power amp design, a pair of Mr Thagard's A100's is doing nothing in the attic of the barn.
Compared to how little the cost was of my 630/9630 A-75 versions i'm likely the biggest twit of all.
Please shoot me, Bob.

[jgeissin]

My thanks to everyone for their advice.
My 'listening area' is in my basement, below grade, so sinking some heat will not be an isssue!

I would LOVE to go into production with something like this, but that is not ethical; at least not without Mr. Pass and Mr. Thagard's permissions and royalties paid back to them. The hardest part would be matching output devices; I need to get someone to make a computer interface to the testing rig so that it can be automated somewhat; possibly even give/plot a curve. But that is outside my knowlege base.

I am in child-like anticipation of getting these completed and listening to them; this is going to be awesome!

Keep the advice coming, I look forward to it! Pics of construction will be on my personal site as it progresses.

Many thanks!

[jupiterjune]

Keep the advice coming, I look forward to it! Pics of construction will be on my personal site as it progresses

Well, I don't know if you found the exact transistors for the front end PS regulator, but a search in Mouser turned up a high frequency driver that I paid $6 a piece for. I think your TO220 output devices would work fine there, or some cheap darlington power transistors. I think I used TIP33c and its complement.

I beefed up the cap values in the front end PS -- the voltage doubler causes a lot of ripple.

JJ

[BobEllis]

You can substitute MJE15032/33 for a lot less than $6. MJE15034/35 or the unleaded versions (MJE1503xG) will work too.

The IRF540s have a very high input capacitance compared to the IRF510/9510 specified for the front end. They are similar to the output devices specified, but you should use the 510 or 610 versions for the front end.

Run through the thermal calculations NP describes in the article for your output devices. You might find that you want to increase their number to keep the junctions within limits.