Another interesting salvage from UNSW – pulled from a disposed machine in Electrical Engineering. This one is a Unisearch PC-ECG v2.1 ISA Card.
As one can guess from the name, this card is meant to perform electrocardiograph functions on a computer. This is another board with almost no information online – so I haven’t really got any idea how to make it work. The serial number of 25060 suggests these devices were somewhat rare and special research devices.
I did stumble across a few papers relating to ISA ECG (although from first glance, this doesn’t appear to be based on that design) and also DIY ECG designs (meaning it’s probably not so complex), but nothing specifically related to this particular board.
Maybe we should just look a little closer – observations:
- Black painted backplate and DB-15 connector with no shell – it appears they’ve taken to great lengths to isolate the PC ground from the measurement probes possibly for safety and noise reasons. Even the “screws” that normally hold the backplate to the PCB are instead plastic nuts of some sort.
- Much of the card (amplifier) components are replicated – it appears it may be an eight-channel device.
- There’s a bunch of DIP switch for configuration, but no clue as to what settings are modified (possibly I/O address, IRQ, DMA channels).
Here’s the plastic ECG connector. Lots of diodes nearby, and Burr Brown INA118U Instrumentation Amplifiers dated Week 8, 1995. the Maxim chip is an IH5042 CMOS Analog switch – maybe this performs a channel selection function. The diodes may be forming a “crude” AND feature for enabling channels – just a suspicion. The H11N2 chips are high speed logic optocouplers – so isolation seems to be a key design/safety feature.
Across the top edge, we see the venerable LM324 Op-Amp. Again, we see CMOS switches in the form of the ADG212A. So many switches! The IL300-F is another optocoupler – so maybe there’s two sets of channels. Adjustment pots are precision 10-turns type with what appears to be liquid paper to “immobilize” them and protect them against vibration. No idea what the 2274ACs are.
Here’s a row of MAX292 Lowpass filters – likely to filter out high frequency noise from the low-frequency sub 50Hz ECG signals. There are also nice golden standoffs for test-points.
In ther corner, there’s a Burr Brown 1.5W unregulated DC to DC power converter module (Burr Brown, more famous for audio Op Amps – no idea they made these). This is flanked by some surface mount high speed CMOS gates.
The core logic of the board appears to be implemented in a PIC, along with an AMD Programmable Array Logic chip. The DIP marked X1 is an oscillator, and so is the SPG 8640BN. The square chip is a precision 12-bit A to D converter with a sample rate of about 100kS/s.
Finally, near the ISA connector, it seems that there’s an LT based regulator U59 and a relatively big ghtrough hole resistor, and two capacitors. The resistor stands proud of the board … one has to be a bit careful of that.