Technical Assistance/Advice Thread

[Pars]

Novice question: 
What is the correct way to tin this type of pad? Should I fill with tin or just put a little under the component as I normally do?
 

 

That looks right to me, but I’m no SMT expert.


Sent from my iPhone using Tapatalk

[mypasswordis]

The best way is to use solder paste and a hot air gun. Not sure how well it would work with a normal iron and solder but definitely don't flood the pad before putting on the transistor. 

[UFN]

I normally solder one of the outer legs on its pad while holding the transistor with tweezers - then it's easier to correct the alignment on the pads if something's wrong. Then I solder the other legs manually and lastly I solder the tab to the pad for good thermal contact. The hot air gun is probably easier, but this works well for me.

[jose]

Thanks guys.

I usually solder smd with a "normal" size but I never solder any with a long legs. 

I tried with a 60w iron (a few second), solder paste and I guess that I have good result.

 

 

 

[kevin gilmore]

that is a bit too much solder on the tab, but it should be fine

[jose]

Thanks Dr. Gilmore !!

[Pars]

Wasn’t quite sure where to put this so...
Ran across a thread on diyaudio for an all tube I/v stage for an ESS9038. I haven’t looked at the chip, but I thought most of the ESS dacs were either v out or could be configured for either?
At any rate, I would suspect this is good in a not good way

Pure Tube i/V for ES9038PRO-9028, 9018, AKM...

http://www.diyaudio.com/forums/showthread.php?p=5259820


Sent from my iPhone using Tapatalk

[cspirou]

So ive been using a toaster oven for reflow. I did a test and the result were promising. I was a little too confident and soldered the QFN chips for the Dynalo mini. First time I used a little too much paste and ended up with something like this.

the positive regulator is out of alignment and the negative regulator has the bridges. I decided to reflow a second time and nudge the chip into place. Now it looks like this.

i thought my problems were solved but when i powered up the positive regulator I only meaured a voltage of 7.87 V instead of the expected 14.4V. This means that only one of the 6.4V pins soldered. Very frustrating.

So what I am wondering is how many times can you reflow a board before the components are affected? Should I remove the capacitors first? I would prefer to use a hot air station but I dont have one.

[Pars]

Can you run an iron w/small tip along the pins, maybe adding solder? Clean the iron good then run it along again. Repeat as necessary until no more bridges?

That would be easier without the two or three caps in place, but probably still doable. Have you got decent magnification?

[cspirou]

The soldering iron I use has dual irons (one fine, one chisel) so removing the capacitors isnt so bad. I assumed I couldn't feed solder because the pins are under the chip but I'll try it out.

[Pars]

Check this post:

 

[mypasswordis]

If you don't want to use a hot air rework station then you're going to need a stencil of that package and to make sure the oven is flat. Hot air station can be had for like 50 bucks

[cspirou]

32 minutes ago, mypasswordis said:

If you don't want to use a hot air rework station then you're going to need a stencil of that package and to make sure the oven is flat. Hot air station can be had for like 50 bucks

Believe me, I learned my lesson. If I attempt soldering this sort of chip again I'm using a stencil. I found them for $10.

http://www.kr4.us/QFN-20-0.65-mm-pitch-5-x-5-mm-body-3.1-x-3.1-mm-pad-Stainless-Steel-Stencil.html

Has anyone been successful applying the iron directly to the heatpad like in this video?

 

[Pars]

[whisper] Ditch Hakko... buy Metcal...[/whisper]

Quoted the wrong post above. This person sounds like they know what they are doing in soldering packages like this.


Sent from my iPhone using Tapatalk

[luvdunhill]

Re-reflowing a part is very hard. Sadly, I have thrown out more than a few of those same regulator packages. Once the board isn’t completely flat it’s very hard to get things rectified.

[ComradeTexan]

Long time lurker, first post here

Did I mess up my Stax 727A feedback mod? I think I'm okay, R21 and R22 are in series, correct?

 

As long as I didn't tear off an essential trace where the top of R22 was I know where to go next. 

I have the old boards with the resistors glued on I figured I might destroy that resistor during the removal, and I have a replacement already on the way. Any tips on dealing with the glue on the other three I still have to pull? 

[ComradeTexan]

Wait of course they're just in series. I can see that on the other side! I'm going to go do something else for awhile

If anyone has any ideas to deal with the glue, I'm all ears still. It would be a lot easier to just cut the resistors in half and worry about making it look nice later

[ComradeTexan]

My 727A mod is successful

I dealt with the glued pieces by digging underneath them with an xacto blade, then broke them off a half at a time. Just getting the first one took longer than the other three combined. It was probably about 90 mins of actual work. 

I'm trying not to worry too much about the differences and trying to just listen and enjoy it for now. It takes me anywhere from two weeks to to two months to truly evaluate gear, and rushing usually does the opposite of help xD

[Tinkerer]

I got lucky recently looking for large toroid transformer magnetic shielding stuff and got a huge surplus reel of mu metal transformer core ribbon for cheap. But after looking awhile on the net, I'm still not quite sure how to use it.

It seems like you need multiple overlapping enclosures for good magnetic shielding, so do I just cut single circumferences and layer them up instead of wrapping one long ribbon? And unlike electrostatic shielding, you don't add a ground to it correct?

Any special heat or lamination concerns?

Edited January 7, 2018 by Tinkerer

[Craig Sawyers]

What is your application that you need mumetal shielding? The stray field from a big toroid is pretty low in the first place.

Anyhow, provided the ribbon is thin you can gently bend it on a large radius  without losing the properties. What it does not like is being bent sharply.

Plenty on the websites of mumetal suppliers about all this.

[Tinkerer]

The Circlotron has long power leads that pick up noise. I already RF shielded and grounded every major length of wire. Noise at this point is mostly from bringing the center toroid up of the three required for the amp, and I'm pretty sure it's magnetic. Basically, if the amp wasn't so massive, it wouldn't be a problem. But it is, so I'd rather take some time to slap metal on things before I have to do some more extreme case mods. Will probably have to separate the transformers more from each other in any case but might as well try the easier stuff first.

 

What I'm trying to understand how you do the shielding itself though. Most places just have a single layer example. And other places sell dual layer bands and such with minimal overlap. I got an eight pound reel of continuous 50% nickel ribbon (Metglas 2826mb). I know multiple layers and thickness improve the shielding by a lot. What I don't get is if you use a bunch of separate wraps or just keep wrapping a single length? Or if a whole bunch of unbonded layers might hum like a loose transformer core? I just haven't seen any of that. I know when you make a transformer core, you use a single length and have all the layers epoxied together for strength and heat transfer. But I have no idea how hot shielding and stuff gets or what it needs to be connected to either. Seems like with mu metal you just make sure you don't bend it severely or solder directly to it or cause a shorted loop, but that's all the detail given.

Edited January 7, 2018 by Tinkerer

[Craig Sawyers]

Formulas for shielding effectiveness here  http://mumetal.co.uk/ . They also do a standard range of mumetal deep drawn and hydrogen annealed transformer cans.

[luvdunhill]

Can you minimize the noise by rotation where the lead out points away from anything sensitive?

[Tinkerer]

6 hours ago, Craig Sawyers said:

Formulas for shielding effectiveness here  http://mumetal.co.uk/ . They also do a standard range of mumetal deep drawn and hydrogen annealed transformer cans.

So separate shielding layers are multiplicative when insulated from each other. That answers my question perfectly.

 

4 hours ago, luvdunhill said:

Can you minimize the noise by rotation where the lead out points away from anything sensitive?

I tried rotating the smaller ones and nothing much happened. Can't rotate the big one as there isn't enough play in the leads, but it never made noise when running the amp just as a Carbon. Kinda wish I had a Gauss Meter with a probe or something so I wouldn't have to do so much guesswork.

[Mr.Sneis]

Hot air rework needed on a pmd200...

 

Any advice for a guy who's just OK with a regular old iron?

I'm local to a shop called circuit specialists that sells some lower end tools like this one for about $60:

https://www.circuitspecialists.com/csi8786d_digital_hot_air_rework_station_with_soldering_iron.html

Also frys locally has super chintzier looking stuff for $100:

https://www.frys.com/product/6389401?site=sr:SEARCH:MAIN_RSLT_PG

I do have a large supply of junk older gear to practice on as well thankfully.

Edited February 14, 2018 by Mr.Sneis