From grh@hinton.demon.co.uk Sun Nov 19 14:19:16 1995 19 Nov 95 17:18:38 +0500 18 Nov 95 08:06:09 +0500 18 Nov 95 12:58 GMT 18 Nov 95 12:58 GMT Date: Sat, 18 Nov 1995 13:01:51 +0000 From: Graham Hinton Subject: Re: HOT Prophets There are two reasons for most Sequential Circuits products running hot and both can be fixed, but should only be attempted by those with reasonable electronics experience. 1) The technology of the day was NMOS and LSTTL and most of the logic chips have modern CMOS and HCMOS pin equivalents which run on a lot less power. Simply replacing the old chips in the microprocessor circuitry will reduce the power required on the 5V rail. If you see a chip marked 74LSxxx it can be replaced with 74HCxxx or 74HCTxxx where xxx remains the same. The Z80-CPU can be replaced with a Z84C00xxP where xx is the top speed and is not critical. Most of the chips are quite cheap, but pick the warmest first and do them a few at a time checking that the instrument is still working in between. Rev1 and Rev2 Prophets used 3-rail EPROMs which are not easy to adapt, but later instruments are quite straightforward. 2) Most of the heat is caused by the inefficient power supply design. SCI tended to use a single transformer secondary and bridge in common for all the power rails. This means that whatever power the 5V rail takes, about THREE TIMES as much is being wasted as heat on the heatsink. Solution: either replace the transformer with one having secondaries suitable for +/-15V and 5V supplies and add another bridge OR just disconnect the input to the 5V regulator and put in another transformer, bridge and electrolytic cap. for that, leaving the rest in place. There is enough information above for any competent technician to attempt the job. Conversely, if there is any doubt don't let them touch your instrument. Stay cool and live longer, Graham Hinton Hinton Instruments Analogue & Digital, Hardware & Software Design for MIDI and Pro Audio WWW Page: http://www.hollis.co.uk/hinton.html From vance@netcom.com Mon Nov 20 06:23:54 1995 20 Nov 95 09:23:52 +0500 19 Nov 95 21:27:29 +0500 id SAA07786; Sun, 19 Nov 1995 18:19:59 -0800 Date: Sun, 19 Nov 1995 18:19:59 -0800 From: vance@netcom.com (Vance Gloster) Cc: analogue@hyperreal.com Subject: Re: HOT Prophets 1) The technology of the day was NMOS and LSTTL and most of the logic chips have modern CMOS and HCMOS pin equivalents which run on a lot less power. Simply replacing the old chips in the microprocessor circuitry will reduce the power required on the 5V rail. If you see a chip marked 74LSxxx it can be replaced with 74HCxxx or 74HCTxxx where xxx remains the same. The Z80-CPU can be replaced with a Z84C00xxP where xx is the top speed and is not critical. I do not recommend doing this blindly. One respect that the more modern chips differ in is "fanout". This refers to how many inputs an output can drive. Usually the more modern chips do not have as much ability to sink or source current (this is part of how they manage to use less current). Also designers sometimes depend on specific side characteristics of a device. It may be that the designer is using the time delay as a signal propogates through the device. In a more modern device this delay time would be less. In many cases it works well to use the more modern chips, but there are some exceptions. -Vance Gloster vance@netcom.com From grh@hinton.demon.co.uk Mon Nov 20 13:17:26 1995 20 Nov 95 16:17:14 +0500 20 Nov 95 09:25:19 +0500 20 Nov 95 14:08 GMT 20 Nov 95 14:07 GMT Date: Mon, 20 Nov 1995 14:10:59 +0000 From: Graham Hinton Subject: Re: HOT Prophets Cc: analogue@hyperreal.com X-Pmflags: 34078848 >I do not recommend doing this blindly. Who said do it blindly? >One respect that the more modern chips differ in is "fanout". >... >Also designers sometimes depend on specific side >characteristics of a device. It may be that the designer is using the >time delay as a signal propogates through the device. In general this might be true, more so for bad designs, but to where on the Prophets do you think this is applicable? The chips that are safe to replace with CMOS equivalents are the CPU, EPROMS, 8253, 7474 and 74138s. "Fain would I climb, yet fear I to fall." "If thy heart fail thee, climb not at all." Graham Hinton Hinton Instruments Analogue & Digital, Hardware & Software Design for MIDI and Pro Audio WWW Page: http://www.hollis.co.uk/hinton.html From vance@netcom.com Tue Nov 21 08:15:15 1995 21 Nov 95 11:14:09 +0500 20 Nov 95 19:02:49 +0500 id PAA18866; Mon, 20 Nov 1995 15:00:40 -0800 Date: Mon, 20 Nov 1995 15:00:40 -0800 From: vance@netcom.com (Vance Gloster) Cc: analogue@hyperreal.com Subject: Re: HOT Prophets In general this might be true, more so for bad designs, but to where on the Prophets do you think this is applicable? The chips that are safe to replace with CMOS equivalents are the CPU, EPROMS, 8253, 7474 and 74138s. This may be the case. The original message seemed to imply a general principle that this could be done anywhere. I was taking issue with this view. It is not bad design to use close to the full fanout that parts are designed for. Also some older EPROMs have amazingly high current requirements for their inputs, so in some designs 74138s would not be safe to replace if you did not replace the EPROMs too. Usually there is at least one stage of a 7404 in most designs that fans out to a large number of places (often designs used 7404s even after 74LS04s became common due to this fanout requirement). These are usually not safe to replace. I do not have the Prophet schematics in front of me so I cannot comment more specifically. I did not mean to flame anyone. I just wanted to keep some enthusastic but technically limited members of the list from accidentally destroying some old synths that there are far too few of anyway. -Vance Gloster vance@netcom.com