A DIY 500 Watt Bass Head

[beans-on-toast]

Speaking of ferrite, Jensen sells what they call FB-2 and FB-6 beads. Have you used these with any of your builds with their transformers. I've found them effective in reducing RF on sensitive inputs.

 

[Passinwind]

Speaking of ferrite, Jensen sells what they call FB-2 and FB-6 beads. Have you used these with any of your builds with their transformers. I've found them effective in reducing RF on sensitive inputs.

Nope, but in my other builds the transformers were on the outputs. Jensen also sells a little inductor for cutting junk on the output of opamp driver stages, which I have used in my last few SS builds. I'm plenty paranoid about parasitics and always 'scope for that, for sure.

 

[beans-on-toast]

Sweating the details is the fun part. To paraphrase Forrest Gump's mama, building an amp is like a box of chocolates, you never know what you're gonna get.

 

[fdeck]

Very nice work!

 

[Passinwind]

Impressive, as usual. The parametric and variable HPF seem like useful additions. Curious about one thing - what is the reason for transformer-coupling the pre and power amps?

Mostly a "secret sauce" habit I've developed, although some manufacturers pimp it as a safety feature as well. I'll be using the preamp and power amp sections separately at times -- they even have their own line cords and switches. So the transformer may occasionally come in handy in some interfacing situations.

Sweating the details is the fun part. To paraphrase Forrest Gump's mama, building an amp is like a box of chocolates, you never know what you're gonna get.

Yep, very true. Bobby was nice enough to follow up with me in a couple of PMs and it turns out that most of his suggestions were already addressed, to some degree at least. I'm lucky to be working off a designer's manual that specs a lot of things I would otherwise need to measure for myself. Off course, I'll probably measure most of that stuff anyway, since learning how to do that effectively is a big part of my attraction to DIY building.

I already had a fairly solid shielding and heatsinking solution worked up, but I found another stash of aluminum bits and also am adding a custom cut sink from these guys. Ordered yesterday, shipped today. My March 30th deadline feels a bit tighter, but I'll give it my best shot.

 

[beans-on-toast]

I already had a fairly solid shielding and heatsinking solution worked up, but I found another stash of aluminum bits and also am adding a custom cut sink from these guys.

Affordable prices. Quality aluminum. Thanks for the pointer Charlie.

 

[Passinwind]

A little progress this weekend, installed +/- 15 volt power supply for preamp section:

This was built up from a kit, available Link Removed. I found the transformer on eBay. On to the preamp boards...

 

[MuthaFunk]

Nice work! Can't wait to see this all come together!

One question:

Doesn't the Ice module have an unregulated aux +/- 24V power supply already?

 

[beans-on-toast]

I've used transformers from Toroid Corp for the power supply in a mic pre-amp. They are good products but I found them expensive. You get what you pay for though. I like toroids because you can easily modify the voltages by adding or unwinding turns. It is easy to do.

One interesting thing is that Toroid now sells affordable kits that come with instructions on how to design secondaries.

 

[Passinwind]

Doesn't the Ice module have an unregulated aux +/- 24V power supply already?

The one I'm using has a +/- 12 unregulated supply, so +/- 10 or so would be about tops for a highly regulated solution using it unless you add a doubler. I really want to have complete autonomy between the power module and the preamp in any case. The board I'm using is overkill though, and I figure I may replace it with one of my own if this project ends up being a keeper.

I've used transformers from Toroid Corp for the power supply in a mic pre-amp. They are good products but I found them expensive. You get what you pay for though. I like toroids because you can easily modify the voltages by adding or unwinding turns. It is easy to do.

One interesting thing is that Toroid now sells affordable kits that come with instructions on how to design secondaries.

This one was about 15 bucks on eBay (and more like $70 from the factory), but it didn't include the mount kit. I potted the center with a threaded insert, so I have around 30 bucks into it. I looked at the kits on their website and rolling my own at some future point is on my radar. I want to get into SMPS design eventually too though. LTspice is heavily oriented toward doing that, so I think I could swing it.

 

[bongomania]

Super cool to see this, as with all your projects. Looking forward to hearing more as it develops!

 

[1n3]

... what is the reason for transformer-coupling the pre and power amps?

Mostly a "secret sauce" habit I've developed, although some manufacturers pimp it as a safety feature as well. I'll be using the preamp and power amp sections separately at times -- they even have their own line cords and switches. So the transformer may occasionally come in handy in some interfacing situations.

Thanks. This project is well over my head, but I still enjoy watching it.

 

[Passinwind]

Just have to wire up the preamp section and I should be making some noise in the next few days. Pics to follow as things progress...

Heh, long couple of days. It took around 4 hours to wire and debug the main preamp board, which had one ghost in the machine, probably caused by corrosion on the nice 30 year old NOS Noble stepped pots I'm using in a few spots. Chemical cleaning didn't get it, so I resorted to filing off the tarnish thoroughly and everything looks very stable and quiet on the 'scope. I even got all the boost and cut orientations right on the first try...

Yesterday: sussed the switching arrangements and how to make the whole mess somewhat serviceable without complete disassembly. Got it all together and realized that the switching was backwards, oops. Fixed that, wired up the new HPF/PEQ board...high pass works fine right off the bat, appears to follow the LTspice model very well for frequency sweep. Not bad, another 4 hours or so to get to that point.

The parametric section had two design glitches on the prototype circuit board, but revision and rework are a frequent necessity in my world and I thought I had that all taken care of months ago. As it turns out, I did, except for documenting one of the wiring changes that went along with the revisions. So I had lots of fun trying to figure out why my PEQ was a wonky oscillator, if it cooked any opamps, if any wires were broken, etc. A few band-aids seemed to help a little, but clearly there was a fundamental error at work somewhere. I started measuring all sorts of stuff methodically, with no joy. At ten hours in I decided to call it a night, checked in on TB, and took a last bleary eyed look at the model, the original circuit board layout, and the revised one that I'll send off for fab once I work all the bugs out. Usually sleeping on things and using fresh eyes in the morning works out, but for some reason this time a state of delirium was just right.

And now when I look at the resistor in the lower right corner of the the board pictured in the first post, it seems impossibly obvious that I left that lead exposed to allow wiring in one side of the boost/cut pot, which has to be moved there from its original pad, P6 for those keeping score at home. Hey, that was easy!

Off to see if I created any new issues with my mistake...

 

[InsanityAmps]

Progress! Design flaws are part of the game. I don't think I've ever seen a new design work flawlessly the first time either at work(automated test equipment) or on my own with audio circuits. Tweaking is all part of the fun Keep up the good work Charlie, you'll get her dialed in soon...

 

[oerk]

Wow, nice project! I've had the idea to build an amp myself before, but it's too big of a project for me to tackle. I'd fail spectacularly at the preamp part, most likely.

The way I understand it, the ICEpower module has an integrated SMPS?

Link Removed has nice Tripath amp kits, but they recommend toroid transformers, IIRC. Which wouldn't save much weight compared to a standard class AB design.

 

[cableguy]

Looking great Charlie!

 

[Passinwind]

Wow, nice project! I've had the idea to build an amp myself before, but it's too big of a project for me to tackle. I'd fail spectacularly at the preamp part, most likely.

The way I understand it, the ICEpower module has an integrated SMPS?

The preamp part has taken me many years to dial in, but the last few have worked pretty well over many gigs. The 500ASP module does have an integrated SMPS, but in the end this build won't be super-light, partly because of all the added shielding and heatsinking I'm using. My only expectation/requirement is that it all fits in the 2RU case I already had, and sounds OK.

Progress! Design flaws are part of the game. I don't think I've ever seen a new design work flawlessly the first time either at work(automated test equipment) or on my own with audio circuits. Tweaking is all part of the fun Keep up the good work Charlie, you'll get her dialed in soon...

Thanks Nate. My old day job gave me a much different perspective on prototyping and I actually really enjoy the rework/redesign side a lot more than I used to. One mentor's mantra was "it's not a project until you're bleeding." He meant it both figuratively and literally.

Plenty of bleeding yesterday, with two more flaws found, and a big lesson in the need to document and double check board specs much better than I managed this time. I lucked out and the rework didn't require any circuit board surgery, which would be a hassle given all the ground plane coverage on this board, another experiment that may or may not make the final grade. Once I made the physical build actually match the SPICE model everything appears to work very much as expected. The parametric EQ's base frequency at knobs-at-noon and sweep range look really close to the modeled behavior, and would probably be even closer if I measured actual individual component values and threw those in the model. It looks like component tolerances may have skewed things favorably, if anything. Now on to actual listening tests and things like detailing the grounding and shielding schemes. I'll post pics of the preamp section later today -- not nearly as pretty as I had hoped for, but that's what rev 1.1 is for...

I'm still hoping to get a bunch of user feedback (aka Reality Check) next week at the Seattle GTG, then on to board revisions as needed.

 

[Passinwind]

Spent the afternoon finishing up the heat sink/shield assembly, fitting a new boost/cut pot, and finalizing circuit board mounting details. Here are the boards in "service mode"; they fold out with the pot leads on the HPF board on top and the preamp board's leads on the bottom. Most of the shielded signal leads are routed between the two boards. For that I used Mogami console cable that I had left over from another project. The pot wiring was done with some nice teflon/silver plate wire I found on eBay, which can tolerate quite a bit of manipulation without breaking.

On to bench and sound checking the assembled preamp section later today, then hopefully the whole thing tomorrow or Monday.

 

[oerk]

Looks professionally done!

I always use whatever cable I have laying around, it never occured to me to buy "nice" cables...

 

[beans-on-toast]

The pot wiring was done with some nice teflon/silver plate wire I found on eBay, which can tolerate quite a bit of manipulation without breaking.

So far so good Charlie. The stepped pots are a nice touch.

I buy my teflon wire and shielded wire from Steve at Apex Jr. He has some quality surplus stuff at good prices.