Earliest example of surface mount PCB?

Thread Starter

graybeard

Joined Apr 10, 2012
11
I have repaired and calibrated vintage Smiths (AKA Jaeger) tachometers from British and Swedish vintage cars of the late '60s & early '70s for a long time. Until recently, I have been unable to repair the third generation Smiths tachometers such as those found in the early '70s V12 E-type Jaguars because they used a TI MIC7/C integrated circuit, which hasn't been available for many years. Here is the schematics for the Gen 3 Smiths tachometers:

1585431446470.png

Recently an alert customer noticed that an obsolete, but available IC called SAK215 had the same pinout and function. I thought it was worth a try to see if it would work in these tachometers. Since the only sources of the SAK215s that I can find are via eBay from China, I initially ordered 5 units to play with prior to committing to a larger inventory. After the first 5 units arrived, I breadboarded the test circuit in the data sheet, but was unable to get that circuit to function.

Fortunately, another customer had a rusty old Gen 3 tachometer from an early 70s MG that he was willing to sacrifice for the cause. So I removed the innards from the tachometer, mounted them to a base, removed the IC and replaced it with a ZIF socket on top of a modified wire-wrap socket. You can see that the test setup worked with the TI IC by the needle position in this photo:

SAK215TestJig1.jpg

Notice that the PCB is glued to the metal base plate. I found this PCB to be very interesting. While the schematics for the E-type Jaguar tachometers are almost identical to the MG tachometers, the E-type PCBs are traditional through-hole PCBs. But the MG PCBs are surface mount. That surprised me completely. I had never seen surface mount PCBs from such an early date.

I also noticed that there were only capacitors on the PCB, no resistors. With a little looking, I figured out that the resistors are actually fabricated into the PCBs. They are black rectangles visible on the top of the PCBs. You can also see traces on the PCB. Here are some photos from different angles that show the pads, traces and resistors of the PCB:

Gen3PCB1Annotated.jpg

Gen3PCB4Annotated.jpg

Gen3PCB2.jpg

Gen3PCB3.jpg

I found the fact that this PCB design is surface mount from the early '70s and has embedded resistors to be fascinating. FYI, the electrolytic cap was placed on top of the IC and welded (not soldered) across pins 1 & 7.

With this setup, I was able to test all 5 of the SAK215 chips and they all worked. I have ordered a bunch more as inventory for future Gen 3 tachometer repairs. I can do incoming QC with this setup to ensure that my inventory is good.

I thought some of you might find this bit of history interesting. Here's some more information about vintage Smiths tachometers if you want further reading: https://accutach.com/smiths-tachs

Does anyone have any earlier examples of surface mount PCBs or even PCBs with embedded resistors?

Stay well,

Mark
 

MrChips

Joined Oct 2, 2009
21,846
I remember encountering in my travels resistors printed on the PCB.
They resistors were trimmed to the required resistance using lasers.
 

dl324

Joined Mar 30, 2015
11,506
I used to design some hybrids when I worked in an R&D lab in Palo Alto in the late 70's. The capacitors were surface mount.
 

Thread Starter

graybeard

Joined Apr 10, 2012
11
Very cool nsaspook. I should have guessed that the space program would be leading the way.

Mr Chips, In the reference design and in the through hole design in the Jags, there are calibration pots, but the MG tach has a fixed PCB resistor. There is a translucent layer over the PCB resistors. The tachometer had to be assembled before it was calibrated. I could not find any evidence of laser trimming of the calibration resistor. Of course that evidence could be on the bottom of the PCB. I am uncomfortable removing it from the plate for fear of damaging it.

dl324, I started working at Intel in 1976. We used no surface mount anything at all back them. I would have been blown away to see SMT back then.

Thanks for the replies!
 

jpanhalt

Joined Jan 18, 2008
10,221
If you consider "Manhattan style" or "dead bug style" construction to be versions of surface mount, they go way back. Manhattan doesn't require IC's or components to be solid state. Both methods were in use by the time you started at Intel.
 

Thread Starter

graybeard

Joined Apr 10, 2012
11
jpanhalt and bertus, thanks so much for the Manhattan information. I never had time to be a ham, so I never ran across those kinds of fabrication techniques.

dl324, I was in microprocessor group marketing for many years in rented buildings early, SC4 later. I was with Intel from 1976 to 2006. I had more fun than anyone has a right to have. :)
 

dl324

Joined Mar 30, 2015
11,506
I was in microprocessor group marketing for many years in rented buildings early, SC4 later. I was with Intel from 1976 to 2006.
I worked on many microprocessors from P1-Haswell.
I had more fun than anyone has a right to have.
I really enjoyed my early years at Intel.

I missed the values and work ethics of Intel in the 80's. The company changed after the mid-90's and is no longer recognizable to me.
 

Thread Starter

graybeard

Joined Apr 10, 2012
11
I worked on many microprocessors from P1-Haswell.
I really enjoyed my early years at Intel.

I missed the values and work ethics of Intel in the 80's. The company changed after the mid-90's and is no longer recognizable to me.
Unrecognizable to me too. :(
 
Those of us who are old enough and fortunate to have worked in a corporation during the 1970s and early 1980s, know exactly what you mean. Well known corporations are no longer recognizable.

I will stop that comment right here, and focus on the OP’s original question.

Printed resistors have been used for perhaps 50 years, in the form of hybrid circuits in alumina or berylia substrates.
Usually for the aerospace and military markets, although I remember that some audio power modules, from ILP in the UK, also employed hybrid construction.
 

Thread Starter

graybeard

Joined Apr 10, 2012
11
schmitt trigger, I just heard from a retired Smiths service technician in NZ that those PCBs do, in fact, contain beryllium. And Smiths is a UK company. I wouldn't be surprised if the PCBs came from the same source as the audio power module PCBs.

You screen name reminds me of one of my EE professors, Dr Schmitz. He had lost the tip of his right index finger to the first knuckle. As a gag gift, we modified a toy gun to turn the trigger sideways and gave him a gun with a "Schmitz Trigger". Fortunately, he had a great sense of humor and loved it. :)
 
Top