• These commercial threads are for private transactions. diyAudio.com provides these forums for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members, use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

New Audio Op Amp - OPA1622

Kamis, that's good information but with this very tiny device the hot air will heat the silicon to almost the same temperature as the contacts. Same with hand soldering. I have hand soldered many SOIC devices, mostly for instrumentation, and all have survived without degradation. But SOIC is huge compared with this tiny 3mm x 3mm device. The OPA1622 might still work after being cooked but it will most likely be degraded. That's not what we want. I think vapour phase is the only way to go. Possibly infrared but it must be done properly.
 
Yes, it is not that difficult to solder these parts given (1) a steady hand, (2) soldering iron with tiny tip and/or hot air tool and (3) good solder. As Chris indicates, a large-ish via can allow soldering of the pad after the leads are soldered. Another approach is to use the tiniest dab of Arctic Silver thermal compound to provide the thermal connection between device and the pad on the PCB and then solder the leads. Just be sure it's a tiny dab and that you don't slide the part around.

One should be able to solder the IC without damaging it.
 
Honestly, we assemble most of our in-house demo PCBs for this part using an inexpensive Weller pyropen hot air torch and some solder paste. Granted, it takes some practice. Occasionally you get a solder bridge between pads which has to be corrected by applying some flux and touching the bridge with a fine tip iron, surface tension does the rest.

After assembly the boards are cleaned and distortion performance checked on the bench to confirm proper functionality.

In a couple months we should have a web-orderable evaluation module available on the eStore.
 
If anyone is interested, I've been authoring a blog series with one of the designers of the OPA1622 that highlights some design challenges of headphone amplifiers and how we addressed those challenges in the design of the part. It will be a 4 part series, parts 1 and 2 are online now:

Part 1: https://e2e.ti.com/blogs_/b/precisionhub/archive/2015/12/21/amp-up-your-cans-how-much-power-do-your-headphones-need-part-1

Part 2: https://e2e.ti.com/blogs_/b/precisi...-op-amps-for-headphones-cap-load-drive-part-2

Part 3 will be online in February and Part 4 in March. Because of the word limit TI puts on blogs, these don't really go as deep as I would like, but it should be good information for some.
 
Member
Joined 2011
Paid Member
Johnc124--

I think I might make up some little bitty SMD-to-thruhole adapter PCBs, solder an OPA1622 onto each board, and sell them to folks here on DIYA who lack the bravery to try it themselves. There will be a markup of course, partly to pay for my time, and partly to discourage volume purchasing of what is clearly a prototyping-only item. A goodnatured fun example might be: N units cost P*(N^1.5) dollars where P is the price for one unit and "^1.5" means "raised to the power 1.5".
 
Johnc124--

I think I might make up some little bitty SMD-to-thruhole adapter PCBs, solder an OPA1622 onto each board, and sell them to folks here on DIYA who lack the bravery to try it themselves. There will be a markup of course, partly to pay for my time, and partly to discourage volume purchasing of what is clearly a prototyping-only item. A goodnatured fun example might be: N units cost P*(N^1.5) dollars where P is the price for one unit and "^1.5" means "raised to the power 1.5".

More power to you! Some tips:

1. If you are trying to map the pinout to the standard 8-pin dual op amp pinout there will be some degradation in performance. Connecting the ground pin to the negative supply will degrade the PSRR a bit. For best performance I would suggest your PCBs just pinout the device directly.

2. You might consider putting the power supply bypass capacitors directly on the PCB so they can be very close to the op amp.

Looking forward to seeing what you come up with!
 
Kamis, that's good information but with this very tiny device the hot air will heat the silicon to almost the same temperature as the contacts.

That's what you are supposed to do. With modern soldering processes every part on the PCB is heated up to the soldering temperature. Parts are designed to cope with this, even plastics and you'll see heating profiles for temperature sensitive parts given in data sheets.
 
Member
Joined 2015
Paid Member
Johnc124--

I think I might make up some little bitty SMD-to-thruhole adapter PCBs, solder an OPA1622 onto each board, and sell them to folks here on DIYA who lack the bravery to try it themselves. There will be a markup of course, partly to pay for my time, and partly to discourage volume purchasing of what is clearly a prototyping-only item. A goodnatured fun example might be: N units cost P*(N^1.5) dollars where P is the price for one unit and "^1.5" means "raised to the power 1.5".

Hi Mark,

Great initiative, I'll order two, and by the way do you plan making same for TPS7A4700 ?....I also may order two.

And thanks to Johnc124, It's very interesting to have the story from inside the chip.

PS : 1,5 power factor is a bit high, have non linear rule : 1 for 1 or 2 sample and 1,5 if more than 2 sample :)

Bests
 
I think it makes sense to build an SMD adapter with SMD components. I cribbed the TI design from figure 60 in the data sheet into this which measures 0.750x0.750 inch:
 

Attachments

  • opa1622_3d.png
    opa1622_3d.png
    41.7 KB · Views: 1,159
jackinnj - looks good!

I was studying that datasheet layout example a few days ago and was wondering how V+ and V- would get to those two interior (inside the ground pour) pins. Are those vias under the two SMD pads to change layers and bring the power rails out? I haven't seen a via used as a SMD pad before. I was thinking that one way to get those power rail traces out was do a right angle and run them under the SMD resistor and capacitor, then out between ground and the V-/V+ traces.
 
Last edited:
You could do a via-in-pad or a via under the body of the resistor to bring the supplies in. Just make sure the supply line runs to the capacitor pad first before going to the chip to maximize the effectiveness of the cap. You could also run the lines under the resistor bodies but that will definitely limit trace width depending on the size of SMD components you're using.

Jackinnj - That looks really good! And will be very useful to a lot of people here. Because your components are larger than the 0402s I used in the layout example in the datasheet, it might make sense to re-arrange the components in the feedback path and the resistor connected to the negative input to try to reduce the area of copper at that op amp input. Keeping parasitic capacitance at the inverting input of an op amp low is always a good goal in op amp layout.

Really excited to see how this turns out!
 
Member
Joined 2011
Paid Member
We hand assemble/replace EPAD's all the time using a temperature controlled hot plate.

Scott, thanks very much! That sounds like a simple and fun way to get the job done. Amazon sells some temperature controlled hot plates for about USD 250 and upward. But I decided to put together my own rig-of-the-merde instead, using

this $15 hotplate

and

this $39 temperature controller.

I hope they work together sufficiently well to approximate a purpose built temperature controlled hot plate, and to solder EPAD devices. If so, big money has been saved. DIY tinkering enjoyment has also occurred!

Amazon talked me into purchasing an optional accessory, not because it is needed, but instead because it looks like FUN: a $14 non-contact (laser) thermometer