INMC News

before shipment. The documentation was accurate, and contained all the information required to get the Imp up and running, with the 100 odd sheets of fanfold perforated paper also provided. The documentation also contained complete software listings of the operating system, and circuit diagrams, although the style of the circuit drawings left a lot to be desired. The documentation also gave a listing to use the handshake with a Nascom, and this was tried and worked. Close examination revealed that the mains fuse fitted (500mA) did not tally with the circuits (850mA), and cases of fuses blowing without reason have been traced to this cause. Internally one screw (on the head guide bar) was found to be only finger tight. Everything else was in order.

A bit of Naspen text was tried, and the result was most acceptable, the character set being well chosen, given the restrictions of the 7 x 7 dot matrix with no decenders. The only confusing characters being the lower case ‘s’ and ‘g’. However, it was noted that the printer was only printing in one direction. This seemed odd until it was discovered that that software counts the characters per line (it doesn’t count spaces) and if there are more than 40 printable characters on one line, then the printer does not print in the reverse direction. This allows time for the print head to cool down between lines.

Next a bit of Basic listing was tried., and, as the lines were shorter, bidirectional printing occured on almost all lines. It was then that it was noticed that the characters did not line up in columns, but that the characters printed in the right – left direction were displayed about half a character width to the right. The cause was found to be the right hand limit microswitch. It seems that the software starts counting the character positions as soon as the switch is released by the print head, and as this is a mechanical device with a fiddly and critical adjustment, this was incorrectly set at the factory. It took about ten minutes to get this switch correct (which probably explains why it wasn’t done properly in the first place), but the result was well worth it! Perfectly regular columns.

The Imp will take rolls of paper (telex rolls are very cheap), sheets (which are little awkward to feed), and perforated paper (which is trifle more expensive). The results were impressive at all times, and the printer averaged about 50 characters per second, bearing in mind that the print head had to do a full traverse even if only one character was printed on a line. The noise level was acceptable in a domestic environment, although aggravated by a resonant living room table.

At the price, the printer is probably beyond the reach of the pockets of many who would like one, but at the price should make some of the big boys sit up and take notice. The Imp is small, light and mechanically and electrically sound (although a little tweaking won’t go amiss), and certainly very good value for money.

As the whole thing is software controlled perhaps we can look forward to software selectable character sizes, and perhaps even graphics. All of which could be achieved with suitable rewriting of the software. Nascom warn users about against having a go at the software, and we endorse this, as any mistakes in the software are likely to be very expensive.