Jump to content

Recommended Posts

Posted
16 minutes ago, naamanf said:

Nate - Amazing work. I’m slowly cutting my teeth on Fusion but so far really like it. 

I'm incredibly slow and inefficient at it which is frustrating given how well I know 2D CAD.  Still it was satisfying to design even a simple part and have it fit.  There's no way on earth I could have designed the actual dust extractor, or put differently I probably could have it would have just taken a year.  But all the required parts are printed now and the three main pieces of the body are permanently affixed.  I'll get one of my homer buckets setup as the first extraction vessel and hopefully make some wood chips next weekend.

 

  • 3 weeks later...
Posted

Tonight was a return to non-wood based making.  One of the other folks retiring this year (good lord, please make it stop) is one of our electrical engineering techs who's just a cool guy.  He's already retired once from being a full time electrician but has been with us for 5 or 6 years.  He sits outside my office so we talk quite a bit and I wanted to make sure that he had a unique way to remember the island when he left.  So I worked with our CADD department to isolate the AC electrical grid that runs underground on the island that Tom (the retiree) had to work with every day. I reverse engraved it on an acrylic sheet and also used the laser to cut it out before mounting it to a lighted LED base (it came as a kit).  Pretty simple project, should make a fun light to have around the house.

FD6FE735-7D9B-45F8-AE79-EA1CDA3313DC_1_102_o.thumb.jpeg.a18afae1e44aea333059639b95490d09.jpeg

The worst part of it was polishing out all the fine scratches in the acrylic that were left by trying to remove the burn-off staining that was a result of the lasering.  I'm happy with the finished product and think he will be too.

  • Like 10
  • 1 month later...
Posted

I don't think I posted about this anywhere else but have been meaning to.  I got back to some non-wood, non-electronics making again during the great COVID family crisis of 2022 and addressed one of my lingering frustrations with my pellet smoker, namely the storage of temperature probes when not in use.  I had gotten frustrated enough that I was about to pull the trigger on some not-horribly priced Fireboard branded probe storage thingies when I figured that some glorious nerd had probably put something up on Thingiverse and that I should check there first.  Turns out I was right.  I tried a couple examples and found pay dirt on my second go.

3E80228A-36C2-4DE1-9BA4-DA4DBF7DB270_1_105_c.thumb.jpeg.ca49e68b67e58902a74a57cef878a42f.jpegE7180E7C-C555-4804-9513-E1BBBA83EFC4_1_105_c.thumb.jpeg.b72b54f0fec4a5c6be711c44338db214.jpeg

  • Like 8
Posted

That's pretty damn cool! I literally feel like it's a whole new world. The sky is the limit on home manufacturing and problem solving ideas.

For an old dude like me, it's kind of like the Jetsons coming true.

  • Like 1
Posted
2 hours ago, swt61 said:

That's pretty damn cool! I literally feel like it's a whole new world. The sky is the limit on home manufacturing and problem solving ideas.

For an old dude like me, it's kind of like the Jetsons coming true.

Yeah, I'm with you Steve in that getting my brain to default to just making something instead of buying it (for weird stuff like this anyway) is a stretch and a change I continue to work on.  Still, with printers getting seriously cheap (see Creality for example) doing just a few projects like this would almost offset the cost of the machine, not to mention other inevitable uses.

Posted
31 minutes ago, n_maher said:

Yeah, I'm with you Steve in that getting my brain to default to just making something instead of buying it (for weird stuff like this anyway) is a stretch and a change I continue to work on.  Still, with printers getting seriously cheap (see Creality for example) doing just a few projects like this would almost offset the cost of the machine, not to mention other inevitable uses.

Pretty awesome, Nate. Amazon has made me so lazy, definitely need to transition the brain to making things.

  • 4 weeks later...
Posted

Dusted off the PSU for the Dynahi.  Will try to finish this by next week.  CanJam got me all fired up!  I realized how good my setup sounds compared to some high end stuff there so I figured I'd finish this project.

dhpsu1 (2).jpg

dhpsu2.jpg

  • Like 5
Posted

It's my favourite amp for dynamic headphones, double mono like yours with psu in a separate box. Sounds so good I haven't even bothered to replace the THAT340 with JFETs.

Posted
1 hour ago, mdr30 said:

It's my favourite amp for dynamic headphones, double mono like yours with psu in a separate box. Sounds so good I haven't even bothered to replace the THAT340 with JFETs.

Mine too...  I have a single ended version from the original headwize site with the jfets and a dynalo mini as well.

I have two sets of amp boards, the lil knight version and the that340 version.  I have the parts for both and jfets for both so I'm not sure which one I'll build.  I'm leaning towards the lil knight boards first, we'll see I guess

Posted
44 minutes ago, sbelyo said:

Mine too...  I have a single ended version from the original headwize site with the jfets and a dynalo mini as well.

I have two sets of amp boards, the lil knight version and the that340 version.  I have the parts for both and jfets for both so I'm not sure which one I'll build.  I'm leaning towards the lil knight boards first, we'll see I guess

If it matters, I built the that340 version because I believe it's the latest version of the balanced Dynahi with the SUSY input IIRC.

I did not use the THAT340 BJT though - I use the jfets and plugged them into the DIP sockets. No other changes needed. 

Going by memory here but happy to dig out more details if you like me to.

Posted
12 minutes ago, mwl168 said:

If it matters, I built the that340 version because I believe it's the latest version of the balanced Dynahi with the SUSY input IIRC.

I did not use the THAT340 BJT though - I use the jfets and plugged them into the DIP sockets. No other changes needed. 

Going by memory here but happy to dig out more details if you like me to.

Thanks, I'm sure I'll need the help.  I remember the lil kinght boards are SuSy but were layed out for the LT jfets and had a regulator for the servo power.  The THAT340 boards did away with the regulator and added the THAT340, probably some other changes as well to account for sand that went out of production.  Since I have the parts for both I'll most likely build them and see what I like better.  I figure why not since I'm only populating the amp boards and drilling the heatsink mounts for the second set

Posted

Almost completed original Dynahi PS.   It was supposed to be my project in the Christmas holidays, but the short circuited one Diode leg pattern to the Ground destroyed some semiconductors.   It worked as a voltage doubler surprisingly and the first stage DC voltage came close to 100 V (!) on the negative channel (That's why I put fatty 100V 4700uF Nichicon there to continue trouble shooting without worrying puncture of the condensers.)  It took more than a month to solve the mystery why it doubled the voltage.

Now it works fine.  Thermally connected 10K Ohm Alpha Electronics precision registers assure perfect match of positive and negative supply voltage.

Housing is a popular Hifi 2000 Galaxy, but heavily modified.  230 mm depth chassis is combined with 170 mm depth top/bottom plates, and a large heatsink (bought as secondhand so cheap)  is attached to the front behind the face plate.  Anyway it will be located below my desk and I don't care much about the appearance though.

Next is Dynahi amplifier...  Lot's of metal work are waiting for me.

One thing I learned in this trouble, LM338T isn't as fragile as thought.   It was still alive even after more than 50 V of voltage gap from input to output was given.   However "alive" didn't mean it worked flawlessly.  After that, Vref of damaged LM338 became ca. 1.6 V instead of 1.25 V.  It's still usable when taking that Vref into account.   I confirmed the output voltage is stable, only shifted.  I'll use it for non-critical voltage supply purpose.

IMG_2287.JPG

IMG_2286.JPG

IMG_2165.JPG

  • Like 1
Posted
2 hours ago, ibuski said:

Almost completed original Dynahi PS.

That's vintage 2007 right there.  I have a working 2 channel single ended original Dynahi amp that I recently retired.  It's a great amp

  • Like 2
Posted

To build Dynahi, I need to match the Transistors, so I made a Hfe & Vbe tester.  I made up it as a chimera of several online information about Tr measurement device.  I posted that schematic to DIYAudio since it worked well in LTSpice, expecting somebody can review it.  However later I found that I need to take two protection Diodes out to make it work in reality.   Also the constant voltage circuit taking DUT in the OPamp's feedback loop wasn't as perfect as simulation.  LT1078 was the most accurate OPamp I had, and achieved 0 error at Ic: 85mA mode, but it left 60mV error to the 4.00V reference VRe with low current.   OPA2134 was perfect at low current, but generated more than 100mV error at 85mA of Ic.  Eventually I used old uA4558TC that maintained average error 10mV to 4V reference over the range from 1mA to 85mA of Ic.   Against all effort of parts selection, I ended up manually adjusting VRe to 4.000V on my power supply for each single Tr/FET.

I could find good matched quad for A1358/C3421 and other Tr, but had difficulty for J109/K389.   Then I bought matched J74/K170 on eBay, but one pair wasn't matched as I wrote in HC before.   Finally I could find good matched pair of them from extra 10 pcs. of purchase in Yahoo auction in Japan.  Reliable seller's J74/K170 even non-matched weren't cheap though.  Now I'm ready to populate them to PCB?  Before doing so, I need to prepare heatsinks to fix A1358/C3421.  And the heatsink requires threaded holes to be mounted to the housing.  The positioning of the holes requires fixing all aluminum plates on the chassis.

ltspice_screenshot_hfe_tester.png

IMG_2101.JPG

IMG_2094.JPG

IMG_2092.JPG

  • Like 3
Posted
On 3/1/2022 at 1:16 AM, ibuski said:

Next is Dynahi amplifier...  Lot's of metal work are waiting for me.

Metal work I meant....

Cutting 230 mm depth aluminum panel (t=3mm) to 200 mm, to fit to tight space on my desk.  Started Cutting with a small tool on the photo.  I could achieve only ca. 0.5 mm depth of scratches on both sides.   Then I clamped that on my wood working bench, and hit hard with a hammer.  It worked.  I could make straight, so-so sharp section.  Some burrs can be removed with a file.  I need to paint the shining edge black.    It shouldn't be so visible as it's on rear side.  This should fit to 200mm length heatsinks with Hifi2000 front and rear panels.

Then many drilling, tapping, chamfering,....  Oh I shouldn't forget to peel anodized black coating to have whole chassis plates conductive.

IMG_2277.JPG

IMG_2299.JPG

  • Like 1
Posted
6 hours ago, kevin gilmore said:

thats a great way to break a tap and ruin the heatsink.

get a tapping jig.

Thanks for suggestion.  You mean a tapping jig like this?

https://biggatortools.blog/tag/tapping-jig/

or like this?

https://www.pinterest.de/pin/563442603351835015/

I've been using this old tap handle without a jig and haven't had any problem on soft aluminum holes.   I could be just lucky.   I know, it's sooo troublesome when the tap is broken in the hole.   Key point is to hold the tap accurately perpendicular to the block surface in the first some turns.   So far it worked.  I may fail it in next time....  Don't know how much that jig costs.

Posted (edited)
2 hours ago, kevin gilmore said:

your first link definitely works. But

i have an extremely old version of this

Precise Universal Hand Tapper, Assures Straight Perpendicular Tapped Holes - 265-110 - Penn Tool Co., Inc

and this is definitely works, hard to use in volume applications

Precise Mini E-Z Hand Tapper, Assures Straight Perpendicular Tapped Holes - 390-252 - Penn Tool Co., Inc

Thank you for links.   The first link product looks definitely for professional or semi-professional.    I have to think if those $100 - $250 is compensated by saving $30 heatsinks.   Probably I won't build more than 2-3 amplifiers with block heatsink in my rest of life 🙂   I learned why not many builders use square bar to mount Transistors, and will use L-angle next time.  Then I can use bolt & nut instead of tapped holes.   Accessing to the screw from PCB top is easy....   It was almost impossible to exchange LM338T on the square bar after mounting large condensers on PCB.   I didn't think about any possibility of repair or trouble shooting.

Edited by ibuski
Posted

So long as I use 3 in 1 oil, I've never broken a tap in an aluminum heatsink. And I've done it hundreds of times. Just use the correct sized drill bit, and preferably a drill press for the initial hole.

  • Like 3
Posted

Metal works made some progress on my DYNAHI housing.  You see original 2004 old PCBs.   I could fix all thread holes position then was ready for soldering.  I kept front and rear panel untouched and decided to start something electric.

 Adjustment of output transistors bias current and DC offset was done so-so OK.   I realized that asymmetric heatsink location to the PCB wasn't good design when considering temperature balance between 2SA1358 and 2SC3421.  While heating up, I had to continuously adjust DC offset before mounting OPamp.   Hopefully DC servo works in real operation.

I started checking the sound with cheapo earphones.  Yes, it makes music so loud.  I couldn't turn the volume pot more than 30 degree.   It seems there is nothing critically wrong, so I switched to others.  Grado requires 45 degree and ETHER2 demanded ca. 80 degree.   I selected the gain ratio 7 with 1200 & 200 Ohm resisters, but it's still a bit too sensitive.

Although I wrote not critical, a big problem is Hum.   It's terrible with Chassis ground connected to PCB ground on the PS.   When I lifted it, it's significantly decreased.   Still it's barely audible at high volume position without music though.   I'll work on this after putting everything in the housing.  I wouldn't leave the PCB ground isolated from the Chassis ground.

IMG_2327.JPG

IMG_2322.JPG

IMG_2304.JPG

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.