I have a project that uses a couple of DIP ICs with 600 mil height (in particular, a sound chip and a microcontroller).

Until now, I had the habit of placing all components outside of the chip area so they can be seen on the circuit board and if I place anything inside, it would be wire jumpers.

In order for me to save board space, I thought of using a DIP socket holder and putting components into that then inserting the chip into the socket so no one sees the components.

I understand to pull this off that I need low-profile components which I am using, but what I want to know is, would this create a negative effect in terms of interference (including EMI)? andare there any other precautions/caveats I should be aware of if I move components inside the socket area instead of outside?

The components in question are ceramic disc capacitors, 1/4 watt resistors, jumper wires, and low-profile crystals with a max speed of 20Mhz.

 

In order for me to save board space, I thought of using a DIP socket holder and putting components into that then inserting the chip into the socket so no one sees the components.

Other than not being able to easily probe those components while troubleshooting, shouldn't be any problem.

I have DIP sockets that have decoupling caps installed on the underside. I've also worked on designs that had resistors and/or diodes buried under IC sockets. Sometimes replacing components under the sockets also required removing the socket.

Disc caps might be a tight fit, but you can find smaller form factors that will fit.

It's standard procedure for me to use sockets whenever I can. It costs a little more up front, but makes component replacement easy.

 

Heat from a resistor would be my first concern, mostly because I know how to estimate that and have no idea about the crystal. I wouldn't think twice about the capacitor.

 

Assembly would be another. Depending on the size of the component, ti could be difficult. getting it inside the socket. I suppose you could place the component first.

 

Heat from a resistor would be my first concern, mostly because I know how to estimate that and have no idea about the crystal. I wouldn't think twice about the capacitor.

thats a good point. In that case, I'll limit resistors under the IC to 1/4 watt high value

 

thats a good point. In that case, I'll limit resistors under the IC to 1/4 watt high value

You will likely not get more than 1 1/4 watt under the socket anyway.

 

Do you really want to do that?
It would be like having to pull the engine to change the oil filter.

 

I have a project that uses a couple of DIP ICs with 600 mil height (in particular, a sound chip and a microcontroller).

Until now, I had the habit of placing all components outside of the chip area so they can be seen on the circuit board and if I place anything inside, it would be wire jumpers.

In order for me to save board space, I thought of using a DIP socket holder and putting components into that then inserting the chip into the socket so no one sees the components.

I understand to pull this off that I need low-profile components which I am using, but what I want to know is, would this create a negative effect in terms of interference (including EMI)? andare there any other precautions/caveats I should be aware of if I move components inside the socket area instead of outside?

The components in question are ceramic disc capacitors, 1/4 watt resistors, jumper wires, and low-profile crystals with a max speed of 20Mhz.

I've done it, it works fine. The resistors were 1k and bigger so essentially no heat. I don't remember ever running into a failed 1k or bigger resistor so I don't see a risk to diagnose or repair.

Since it is a socket, you'll be able to pull the chip if you ever need to get to them.

Fire away.

 

My first concern was heat. But after reading the positive comments on doing so I guess heat would not be a serious issue, provided you account for that. My second concern is for solder bridging. Since I've never put a passive under a chip (or carrier) I don't know how big a problem it would be to have solder bridging going on. Once it occurs, the only way I can see rectifying that issue would be to remove the socket, then desolder the bridge. Of course, like I said, I've never done this so I can only imagine possible issues.

Sometimes I can be one of those guys who finds problems with the solutions. I hope I'm wrong.

 

My first concern was heat. But after reading the positive comments on doing so I guess heat would not be a serious issue, provided you account for that. My second concern is for solder bridging. Since I've never put a passive under a chip (or carrier) I don't know how big a problem it would be to have solder bridging going on. Once it occurs, the only way I can see rectifying that issue would be to remove the socket, then desolder the bridge. Of course, like I said, I've never done this so I can only imagine possible issues.

Sometimes I can be one of those guys who finds problems with the solutions. I hope I'm wrong.

Solder bridges would, generally, happen on the copper side. The OP told is in Post 1, that he is using LARGE DIP 0.6" parts. He can get 5 parallel 1/4 watt resistors under there! (with 0.1" spacing!). Solder bridging is no bigger issue whether he sets a socket over his row of resistors or not.

 

I once had to repair a DEC PDP-15 that had been running well for over 10 years and then developed a short across Vcc and GND.

How do you find a short with hundreds of ICs and decoupling capacitors in parallel across the supply lines?
The culprit was a blob of solder sitting underneath a DIP IC that had gradually flowed over time.

Solder still flows with gravity and elevated temperatures.

DEC Flip-Chip boards are always mounted vertically which allowed the solder blob to flow and cause the short circuit. No amount of visual inspection was able to detect this.

The flaw was isolated using an Hewlett Packard current probe.

 

I have a project that uses a couple of DIP ICs with 600 mil height (in particular, a sound chip and a microcontroller

How do you find a short with hundreds of ICs and decoupling capacitors in parallel across the supply lines?
The culprit was a blob of solder sitting underneath a DIP IC that had gradually flowed over time.

So, one problem after 10-years of continuous service on a board with 100s of ICs sounds like a ringing endorsement to move full steam ahead on a board with two ICs that will be used (I assume) occasionally.

So, What audio chip are you using? I posted an old AY-3-8913 above.

 

So, one problem after 10-years of continuous service on a board with 100s of ICs sounds like a ringing endorsement to move full steam ahead on a board with two ICs that will be used (I assume) occasionally.

On the contrary. I vote not to do it.

Murphy's Law will prevail.
A fault will develop in the most difficult place to access.

 

I've never heard of cold flowing solder. I HAVE heard of "tin whisker" growth. Among other things, satellites are not allowed to have tin lead solder because the tin can grow whiskers that can become conducive short points and render a satellite in orbit as orbiting space junk.

http://cave.auburn.edu/rsrch-thrusts/mechanism-of-tin-whisker-growth-in-electronics.html