Amprix Mystery Device

Over the years I have accumulated a lot of surplus electronics. I had a bunch of these small PCBs and I had no memory of where they came from. The only marking on the board is "AMPRIX 3-1391". A Google search for Amprix doesn't turn up much, but I did find a few useful pieces of information. First I found a surplus site selling an Amprix voltage regulator. 

https://www.electronicsurplus.com/amprix-electronics-a118-2-voltage-regulator-assembly-adjustable-24vdc

I also found this site which says this was a company from Texas that existed from 1981 to 1984. 

https://opencorporates.com/companies/us_tx/0054892500

Finally I found this quote:

“Amprix designs and builds electronic components and products for the automobile aftermarkets . “

The board only has two inputs, power and ground, and 4 LEDs, so my assumption was that this was some sort of voltage monitoring device. 

Here is the schematics I traced from the PCB.

The D27 zener diode creates a reference voltage which goes through a resistor network to one input of three LM339 comparators. There is a also a variable resistor, R22, to calibrate the reference voltage. The other input of the comparators goes to the supply voltage. With this setup LED 8 and 20 will come on around 5 volts, 8 will come on alone around 9 volts, all the LEDs will go off around 11 volts and finally 14 will come on at around 13 volts. Based on the mention of automotive components and all the LEDs being off around 12 volts (auto battery voltage), this will likely a device for monitoring car batter y voltage. 

The one mystery with this circuit is LED D20, I have found no scenario which turns this one on. It is hooked to the top LM339. The non-inverted input come from the supply voltage and the inverted input from ground. For LED D20 to turn on the non-inverted would have to be higher then the inverted, and I am not sure how that could ever happen. 

My one theory is that the circuit if actually designed incorrectly. Sine I have a bunch of these they were likely surplus components, so maybe these were defective and sold as surplus. 

Missile Target Display Part 2

In my last post I started doing a more detailed look at the Identification Friend or Foe (IFF) system component shown in a recent video from Fran Blanche.

https://www.youtube.com/watch?v=dXrLSOReMFA

The information in this post is based on only what I could see in the video, so there things I cannot be sure of.

On board A3 there is an array of 12, 4-bit register ICs. Since the displays on the front panel only need 4 bit each, it’s safe to assume that these registers are holding the values for 12 of the digits being displayed on the front panel. 

The A2 board has another set of four registers. These could be the remaining registers for the main part of the display, or they could the mode display on the left. There should still be four more registers somewhere so they may be on the one board that wasn’t shown in the video.

Here is one other interesting circuit I was able to identify, there are three instances of this on the boards I can see. The 7485 is a 4 bit comparator, it takes two 4-bit inputs and outputs one of three signals indicating if the one input is less then, greater then, or equal to the other. We can see that one set of inputs is connected to 7474 flip flops. I wasn’t able to see which of the three outputs is being used. One possibility is that these circuits are determining which data gets displayed on each row of the display. The desired display could be latched into the flipflops and when the corresponding data becomes available the “equals” output would latch in that data.

Missile Target Display Part 1

A while back I ran across Fran Blanche’s YouTube channel. Fran does a lot of  great electronic engineering videos, especially videos about vintage electronics. Recently she did a video about a component from an 80’s era missile tracking system. You can see the video here:

https://www.youtube.com/watch?v=dXrLSOReMFA

The video has some good closeups of some of the boards, good enough quality that I could trace out some of the circuits and learn more about how the device worked. The device she shows is an intra-target data indicator display out of an Identification Friend or Foe (IFF) system. One of the comments on the video provided a link to a document about this system:

https://www.globalsecurity.org/military/library/policy/navy/nrtc/14308_ch8.pdf

The device has four circuit cards that go into a small rack assembly and are connected with a backplane. There is also a circuit board on the front that holds the LED displays. The video provides a good look at the front and back of two of the circuit card, shows the front of the third, but does not show the fourth at all. There is also some pretty good shots of the front panel board. Since these are just double sided boards I was able to trace out a good bit of two of the boards. Of course in cases where traces go under chips I can’t tell where they go, so the circuit diagrams are far from complete.

Let’s start with the front panel, here is a picture from the video.

As you can see there are 20 seven-segment LED displays. According to the document linked above the four on the left show what “mode” is being displayed on the ones on the right. The four chips are Fairchild 9368dc’s which are seven-segment decoder/drivers. A four bit binary value in input into A0-A3 and the chip will output the correct segment pattern to generated the digits 0-9 and the hex letters A-F. The displays also have a decimal point which this chip does not control. I am not sure if the decimal points are used or not.

Since there are only four drivers on the front board, and only 30 wires running from the front board to the rest of the device, it’s not possible for there to be a single decoder chip for each seven-segment display. Based on this they must be multiplexing the decoders so each can drive multiple displays.

Looking at the schematic I traced for the A3 board I found this circuit:

The configuration of the transistors appears to be what is known as a Darlington Pair. This is a way of connecting transistors to allow them to drive a high current device. The resistors are also common in this configuration to bleed voltage from capacitance that is inherent in transistor which allows for faster switching.

Since there are four identical circuits here and four drivers on the front panel, my guess is that this combination drives the 4x4 array of displays. Based on the traces I can see on the display board the driver chips are connected to all four displays in a column, so each of these driver circuits would enable one row at a time thus allowing the four drivers to drive 16 displays. I wasn’t able to trace where the NAND gate inputs went but I assume one input of each goes to a circuit that is sequencing through the four rows of the display. The other input may be used to blank out a row that has no data to be displayed.

So, what about the the first column of displays? That is probably where this chip from the A2 board comes into play.

I assume this one is multiplexed to drive the first column. This is further supported by  the fact that the A2 and A3 pins are tied together, which makes is so the chip can only display 0-3, and C-F. Looking at the document I linked to above it appears that that those displays only ever need to show 1-3 and C so this configuration would support that.

HPC Microcontroller Development System–Front Panel

The final piece of the system to look at is the front panel. Here is the front of it where you can see the reset switch, a switch to control the EPROM programmer voltage and the ZIF socket for the EPROM programmer.

On the back there is a small PCB that connects to the two switches. Most of the circuitry on here appears to be to generate the EPROM programmer voltages. Below that PCB is the connector for the ZIF socket. The interesting part of this is that the whole front panel is actually a PCB itself. You can see where the ZIF socket is soldered in and the traces that lead to the connector.

HPC Microcontroller Development System–Pod2

This is the second board in the pod called the SUMMIT-POD2 board. The blue connectors connect to the buffer board and the part on the right connects to the processor socket on the target system. I will talk about that a bit more below. The board is composed of some standard TTL logic chips, a couple PALs and two large PLCC packaged chips. The U10 socket was empty on the board when I got the system. The top chip is an HPC46003V20, which is the processor that this system emulates. It’s not uncommon in an in-circuit emulator to have the processor being emulated in the interface pod where it can be physically close to the circuit that is using the processor.The second larger chip is labeled SCX6B48AHF. I believe this is a custom gate array chip designed just for this system. I assume this chip provides the interface between the development system and the processor being emulated.

Here is the back of the board. On the right you can see how the pod plugs into the socket on the system under test where the processor would go.

Here is a closeup of the ribbon cable that connects the two boards. The cable is actually manufactured into the PCB. The ribbon is sandwiched into the inner layers of the PCB. This makes the connection between the boards flexible so the pod can easily be connected to the processor socket.

HPC Microcontroller Development System–Pod Buffer

In my last post I showed the buffer board that was in the main unit. The board connected to the interface pod with a long cable which connected to another buffer board in the pod.

The board is composed mainly of bus driver ICs and DIP resistor packs. There are also PAL (Programmable Array Logic) chips. The connectors on the left appear to be for external trigger signals that the system can monitor and react to.

On the back of the board the three blue connectors connect to the other board in the pod, and there for connectors on the left are for the cable to the main unit. There is also two pin connector at the bottom the brings power to the pod.

The most interesting thing about this board is that it is the only surface mount board in the system, everything else is through-hole. I did notice that the chips on the other boards had date codes from the late 80’s up until 1990, but this board has date codes of 1994 and 95. I guess it’s possible this board was originally done as through-hole but at some point in the life of the unit it got replaced with newer board which was surface mount.