Actual Compass XP-PRO main board rev:F schematic

[Davor]

OK, I may have an explanation for this: a method to flabbergast copiers. It worked on me, and I have no intention copying it.
I realised this method may have merit, but of all possible component value combinations the designer chose the most ridiculous one, simply because he could. Instead of R34,R35,R36 values being 430M,100k,33k, exactly the same result would happen with 4M3,1M,3k3.

[kt315]

i remember about 560M disputed but in another compass... http://compass-metal-detector-forum.....html#a7607123
i see this R47 560M on schematic of CS PRO of G. Paine.

[Melbeta]

Compass engineers did many things to complicate and defeat those who sought to copy their creations, and flabbergast copiers, including having chips Mis-Marked to represent other chip structures. So unless you had Compass factory schematics, you just might have been taken around the hill, and through the woods, rather then over the bridge across the wide river, and straight along the correct road to Vahalla! This is why I have a older chip machine to read the structure, even if the print was ground off on top of the chip, and then I knew what the chip structure really was/is. Unfortunately the fellow who used to make the machines no longer does that for some reason. I think I know the reason. They took him to court and broke him of the habit... But I still have the older machine, but it only runs on Microsoft Dos, and so I still keep that old machine, and keep that old Microsoft Dos Computer...
Melbeta

[DavidB]

Most of them used the 560 Meg in latter revs, I suspect it was just what ever high value they were able to purchase at the time. Seems strange using a high v They never filed the patent on this model anyway. I noticed I swapped the E, C on Q5 when I copied it over. Is there away to delete drawings and just keep the corrected ones to keep from cluttering this forum.

[kt315]

Is there away to delete drawings and just keep the corrected ones
---
save a file with same name. i think old file will be changed by new.

[Davor]

I'm looking at the main board...

I think lower side of C42 and R99 should go to -5V.

Q5 emitter and collector should be swapped

U5D and Q3 make an interesting precision peak detector that promotes binary response of a discriminator (which I personally hate), but sadly, masks all weak signals after a strong one, and before it in return pass of a coil. It may work marvellously against chatters, but makes it useless on a trashy terrain. If I was given this detector, my first mod would be disabling this circuit by removing R50.

I'd surely not build this thing (having one device with abysmal recovery speed is quite enough), but it is a kind of fun going through the circuitry and trying to grasp the author's reasoning.

[DavidB]

I sure would not build this 30 year old circuit the way it is, the idea was information so more people could have a repair reference. If you disable U5, Q3 the target will produce a long smear until the filter rings down. You would have to modify the filters a little to use it that way but it would work better in moderate ground.

[DavidB]

C42 and R99 just look funny because they are a factory add on mode soldered across R19

[Davor]

It is not the only odd thing in this detector.

Removing R50 to disable peak detector action on Disc mode would require a little modification of threshold, but surely the detector filters are likely to flutter, simply because there are far too many of them, and they are not detuned at all. That would be my second mod.