Sacrificial work holders for headphone amp project

[homebrewed]

I have designed and built a class-A headphone amplifier (a variation on the classic "JLH" circuit) and need to package it up. My plan was to give it some nice "heft" by using .25" thick aluminum stock, which means I need to drill a lot of accurately-placed holes to screw everything together. For one-off projects I have used 1-2-3 blocks and clamps to elevate the plate, aligning it with a DTI. But the number of holes for this project meant I would spend a lot of time just on the alignment step(s). So I designed and 3D-printed some sacrificial holders to hopefully speed things up.

Here's a photo showing them in use:



Two holes done, many more to come. Here's a photo showing a test fit and how things will eventually go together:



It can be seen that I also have printed some board mounts for the PCBs. More holes to drill (and in this case, tap).

The heat sinks were scrounged during my working years and had no provision for mounting them on anything (they originally were glued onto the back of a high-speed hybrid amplifier used in a 1GHz Tektronix oscilloscope). So my mounting scheme uses those .25" square rails on the top and bottom of the heat sinks. This scheme leaves four blank spaces, two on each side of the heat sink, and they will be filled in with more 3D printed pieces. They will be screwed onto the back of the heat sinks. I got some black PLA filament for them because I'm going to paint the enclosure black (with the exception of the heat sinks). Ye olde rattle can approach.

I also am planning on 3D printing some feet because there will be some screw heads protruding on the bottom. The countersink diameter for #6 flat head screws would put the countersink a little beyond the edge of the base (and top). TPU would be the best in terms of not scratching the furniture but I'm skeptical that my old Ender 3 will be able to print it. Instead, I'll use some of those PSA-backed furniture pads.

 

[Flyinfool]

Nice work.
I often 3D print machining fixtures. It works great.

 

[homebrewed]

Nice work.
I often 3D print machining fixtures. It works great.

Thanks for the comment. I'm a recent arrival to this kind of application so it's been a bit of a revelation on what's possible. The print time can be a few hours per part, but.....I don't have to be there for it to happen, unlike the usual manual milling process. That's pretty nice.

 

[WobblyHand]

Re: TPU, does your printer use a Bowden tube, or? On my MK3S and MK4 I have been able to print TPU if I reduced the feed gear tension to the minimum. Basically the spring is loose. Otherwise I would get jams, which ruined the print. I also found a high flow TPU, Sainsoft GT-3, which worked much better than the cheap Overture TPU. Less stringing. TPU absorbs water, you need to dry it at 55C, not 45. After that, I was able to get decent TPU prints. Printing TPU can be frustrating...

 

[dkemppai]

Audio, and Tektronix mentioned in the same post. You have my attention!

Having a 3D printer, and remembering that you have a 3D printer are two different things. I'm so used to machining things, that I forget that some things can be printed! lol.

One of the deciding factors in buying my first machine tool (mill drill) was to be able to machine and drill heat sinks for amplifier projects. That bug comes and goes, and the machine shop has grown, but that very first mill drill is still being used today.

Glad to see others here playing with audio stuff. I took the opportunity Covid provided to pick up the hobby again for a while (Attached images are the final testing...). Unfortunately, since 2020 not much has happened again. The amplifier is in the living room, although it hardly gets used. What I've learned is I like building them more than listening to them. (And, quality recordings aren't always easy to come by...)

Do you have any means to measure performance of the amp? Maybe a build thread on it???

 

[ttabbal]

Re: TPU, does your printer use a Bowden tube, or? On my MK3S and MK4 I have been able to print TPU if I reduced the feed gear tension to the minimum. Basically the spring is loose. Otherwise I would get jams, which ruined the print. I also found a high flow TPU, Sainsoft GT-3, which worked much better than the cheap Overture TPU. Less stringing. TPU absorbs water, you need to dry it at 55C, not 45. After that, I was able to get decent TPU prints. Printing TPU can be frustrating...

Unless he's upgraded it, the older Ender 3s were all Bowden. TPU will be extra hard to print on a Bowden. Maybe one of the stiffer versions that are around. It's even difficult to run in the Bambu AMS units because of that. I have heard of a few people getting those to run in the AMS, so maybe.

 

[GeneT45]

Unless he's upgraded it, the older Ender 3s were all Bowden. TPU will be extra hard to print on a Bowden. Maybe one of the stiffer versions that are around. It's even difficult to run in the Bambu AMS units because of that. I have heard of a few people getting those to run in the AMS, so maybe.

I have yet to try it in my Bambu (and Bambu's official stance is "no TPU") but I've seen a lot of folks say that it's worked for them. I would say PLA is the least AMS friendly material...

GsT

 

[homebrewed]

Re: TPU, does your printer use a Bowden tube, or? On my MK3S and MK4 I have been able to print TPU if I reduced the feed gear tension to the minimum. Basically the spring is loose. Otherwise I would get jams, which ruined the print. I also found a high flow TPU, Sainsoft GT-3, which worked much better than the cheap Overture TPU. Less stringing. TPU absorbs water, you need to dry it at 55C, not 45. After that, I was able to get decent TPU prints. Printing TPU can be frustrating...

Yep, it uses a Bowden tube.

Thank you for the tips, particularly the drying requirement.

 

[Boswell]

I have no problem printing TPU on my Bambu, but I do bypass the AMS as suggested by Bambu Studios.

 

[homebrewed]

Audio, and Tektronix mentioned in the same post. You have my attention!

Having a 3D printer, and remembering that you have a 3D printer are two different things. I'm so used to machining things, that I forget that some things can be printed! lol.

One of the deciding factors in buying my first machine tool (mill drill) was to be able to machine and drill heat sinks for amplifier projects. That bug comes and goes, and the machine shop has grown, but that very first mill drill is still being used today.

Glad to see others here playing with audio stuff. I took the opportunity Covid provided to pick up the hobby again for a while (Attached images are the final testing...). Unfortunately, since 2020 not much has happened again. The amplifier is in the living room, although it hardly gets used. What I've learned is I like building them more than listening to them. (And, quality recordings aren't always easy to come by...)

Do you have any means to measure performance of the amp? Maybe a build thread on it???

Right now all I have are the LTSpice simulation results, which indicate that the second harmonic is about 90 dB down from the fundamental at 1KHz. But that most likely is a best-case situation, real-world numbers almost certainly will be worse than that.

The main limitation for me is the absence of a good low-distortion source. My cheap DDS based signal generator is nowhere good enough as it is. Gotta work on that.

The design doesn't have huge open loop gain to reduce distortion, instead it uses local feedback. As a result, the circuit is oscillation-free and has pretty good bandwidth without requiring fancy compensation schemes.