Monroe 430 Calculator repair and cleanup

I man­aged to snag a Monroe 430 cal­cu­la­tor that was in good con­di­tion and a rea­son­able price. The Monroe 430 is 14 dig­it four-func­tion cal­cu­la­tor with mem­o­ry, per­cent, and a few oth­er extras. I have worked on a mod­el 425 and 420 in the past so I kind of knew what I was get­ting into, espe­cial­ly the lack of documentation.

The 430 is the sec­ond from the top of the line of the Monroe 400 series of cal­cu­la­tors, which include the 410, 415, 420, 425, 430, and the 440 mod­els.
This mod­el 430 cal­cu­la­tor was built in late 1974, but the mod­el was intro­duced in 1973. It uses the Rockwell A4541PB VLSI cal­cu­la­tor on a chip inte­grat­ed cir­cuit.
Compared to the Monroe 425 cal­cu­la­tor, this mod­el adds two addi­tion­al dig­its, selec­table round­ing, con­stant “K” selec­tor, ∑ / ◊ selec­tor, “RV” reg­is­ter swap, Change Sign, Percent, full mem­o­ry con­trol keys, and Memory LED indicator.

The low­er case is held in place by four screws that when removed allow the top assem­bly to eas­i­ly slide out. While not ter­ri­bly dusty, there was a lot of loose junk inside the case which includ­ed dete­ri­o­rat­ing foam from the key­pad, and hot-melt glue chunks used to secure the dis­play mod­ule.
The dis­play mod­ule is a Burroughs Panaplex II gas-dis­charge mod­ule with 14 sev­en seg­ment dig­its, 15 dec­i­mal point / com­ma indi­ca­tors, and a minus sign. Separate from the Panaplex mod­ule is a + — X ÷ dis­play mod­ule that indi­cate the func­tion being performed.

The func­tion­al parts of the cal­cu­la­tor include five sections:

1 — Transformer and frame assem­bly with AC pow­er input con­nec­tor, bulk fil­ter capac­i­tor, and pow­er tran­sis­tor attached to frame as a heatsink.
2 — Rockwell A4541PB proces­sor and inter­face board
3 — Keypad assem­bly
4 — Front switch and dec­i­mal point slide switch board
5 — Display inter­face and pow­er sup­ply board

The brown card edge con­nec­tor in the above image con­nects to the front switch board.

The proces­sor / inter­face board is con­nect­ed to the key­pad assem­bly and the dis­play inter­face / pow­er sup­ply board using some Litton 32 con­tact card edge jumper plugs.
The plugs car­ry sig­nals and pow­er between the boards.

Other than the proces­sor and con­nec­tors there is not a lot hap­pen­ing on the board, the MC14011CP IC is a quad 2‑input NAND gate, and there is an array of diodes for the key­pad assem­bly sig­nals. The front pan­el LED indi­ca­tors are con­nect­ed to the board using a keyed 6‑pin card edge connector.

There is quite a bit more hap­pen­ing on the dis­play / pow­er sup­ply board. This is where the first repair was required.

I found a dis­col­ored sec­tion of cir­cuit board direct­ly below a TO-92 tran­sis­tor. A quick diode check with my mul­ti­me­ter showed no issues, so I reflowed sev­er­al of the con­nec­tions and added a TO-92 heatsink along with some ther­mal paste to the tran­sis­tor. My best guess is that this is part of the pow­er sup­ply and is a result of many years of oper­a­tion in pos­si­bly a hot environment.

The front pan­el switch board pro­vides a con­nec­tion of the slide switch­es to the proces­sor / inter­face board. It includes the slid­ing con­tacts for the 11 posi­tion dec­i­mal point slide switch, and the three larg­er slide switch­es not includ­ing the pow­er switch which has a sep­a­rate connector.

The key­pad was in the worst con­di­tion, and end­ed up need­ing 9 key-switch­es replaced. The foam around the key caps was very dete­ri­o­rat­ed and was either a fine pow­der or larg­er chunks.

The cir­cuit board that inter­faces to the key-switch con­tacts had a lot of dis­col­oration around the con­tact pad areas. I used a poly­mer eras­er to remove most of the tar­nish and then some DeoxIT D5 con­tact clean­er to com­plete the process.

Even after clean­ing the con­tacts of the key-switch mod­ules quite a few of them were still non func­tion­al. There were also some inter­mit­tent mod­ules that worked for a while but then sud­den­ly stopped work­ing. The assem­bly with it’s 13 screws was prob­a­bly dis­as­sem­bled and re-assem­bled a dozen times before every­thing was stable.

I decid­ed to dis­as­sem­ble a cou­ple of the key-switch mod­ules to see what the fail­ure point might be. When the switch plunger is depressed a met­al spring press­es on the red plas­tic fin­ger which in turn press­es a met­al strip into con­tact with a sta­tion­ary plate con­tact.
If the blue con­tact assem­bly is pressed togeth­er using light fin­ger pres­sure a good con­tact is made, but using the flat spring and red con­tact fin­ger there is no switch clo­sure.
I might try some future mod­i­fi­ca­tions exper­i­ment­ing with the bad con­tacts, but for now the cal­cu­la­tor is work­ing, which was the main goal.

Once again I used my stan­dard RetroBrite pro­ce­dure on the low­er case shell to reduce the yel­low­ing. The case is much larg­er than the mod­el 425 and required a medi­um sized con­tain­er and two gal­lons of 12% hydro­gen per­ox­ide solu­tion to com­plete­ly sub­merge the entire case, which was left in the solu­tion under a sun­ny sky for 7 hours.

The Monroe 430 uses the same 115 volt pow­er cord as the mod­el 425 cal­cu­la­tor. The end is marked with a 168, which I am assum­ing is the plug type.

The cal­cu­la­tor has been work­ing well for the past sev­er­al days, with no over­heat­ing issues or addi­tion­al key-switch fail­ures.
In my opin­ion it is a great look­ing cal­cu­la­tor, and I appre­ci­ate the extra dig­its and fea­tures on this mod­el, espe­cial­ly the mem­o­ry.
I did end up replac­ing both of the elec­trolyt­ic capac­i­tors in this unit, but left all the dipped tan­ta­lum capac­i­tors in place.
I would like to find a users man­u­al for this mod­el as I am inter­est­ed in the uses for the accu­mu­la­tion selec­tor switch.