Side Cover Fan

Model 95 Side Panel Fan (by Jim Shorney)

Side Cover Fan
Side Cover Fan Orientation
Case Airflow
Side Cover Fan Construction
Side Cover Fan Revisions
Cover Fan Cable Assembly
Side Fan Connector
Side Cover Fan Troubleshooting
Deriving 17 V DC from the Power Connector


Side Cover Fan

Part of the following was redone based on info found by Ross Barker (Ross pulled up anchor - he's on the high seas again).

Side Cover Fan Orientation

   This is a swirling debate... For now, I'd like to point out the difference between the 95 and 95A plastic end bracket for the fan. The 95A bracket has a slot and a beefier catch.

Airflow

   The fan serves one purpose in either mounting orientation, that of pulling air IN through the grilles at the top front of the 95, then THROUGH the adapters, then DOWN the side wall.

Open to interpretation... The significant points- air is pulled through the top grilles, sucked through the adapter cards, down the side wall through the side wall fan.

   Air is sucked over the top of the PSU, past the complex, over the memory, then into the PSU where it is blown out through the exhaust. Though looking at a PSU shows the memory has a blank surface in front of it, probably to force airflow past it.

Note: The PSU for the 3511 Enclosure has the louvers on the lower, inner edge taped off. These louvers ventilate the SIMMs on the 95 planar. YMMV, I need to source this factoid...

Side Fan Construction

From Peter:
   But please do not forget that this thing is not just a simple "motor". It is an electronically regulated, brushless DC-motor. The electronics sits under the end facing to the rear - you can see the printboard and some of the coils. The electronic itself consists out of some resistors and condensers, ICs, a hall-generator for measuring the rotational speed and so on.

   Unlike to "real motors" the resistance does not change when you turn the fan... normal DC-motors act as generators once driven. This thing doesn't.

However: 137 Ohms would give a current of about 125 mA - still below the specs.


Side Fan Revisions

64F4470, EC C31557, mfd. 042391, 137 Ω
64F4470, EC C32546, mfd. 121391, 1.31 KΩ
Hosiden HMK 3404-01-092, DC 17 V, 0.185 A

64F4470, EC C32546, mfd. 031892, 1.32 KΩ
Hosiden (flower symbol?) W, HMK 3404-01-092, DC 17 V, 0.185 A

95A Side cooling fan fan
ASM P/N 61G3813, mfd. 101993, 1.32 KΩ
Hosiden (flower symbol?) J, HMK 3404-01-092, DC 17 V, 0.185 A

Possibly this means any fan made on or after 121391 will be a 1.3 KΩ model.


Cover Fan Cable Assembly

HMM says 61G3824, the older style is P/N 84F9284, the newer style is P/N 60G9828. For purposes of clarity, I wedged the white planar connector into the top.

The older one is black, and has a full-height pocket for the inner mounting screw.

The newer one is off-white and the bracket is redesigned to be half as high, but with a loop sticking up high enough to mate with the mounting post.

Screw appears to be a 3 to 3.5 mm by 8 mm long trilobular screw, double pitch (like they stretched the screw and the thread is farther apart).


Side Fan Connector

The fan connector is located on the planar and provides 17 V DC directly from the power supply. It does this by combining voltage from +5 V rail and -12 V rail.

Pinout of the side fan planar connector (male, top view):


Side Fan Troubleshooting

If the access cover fan does not work:
   Check the spring contacts on the fan bracket to see that they stick out far enough. Over time, with repeated removal and installation of the side cover, the contacts will be pushed back into their guides. Carefully pull them out again. Make sure the free end of the spring enters the recess when not under pressure (that way it's lined up when you are using BOTH hands to install the side cover...)

Power to fan: 17 V DC (+/- 1.4 V DC) at the two fan cable pins on the base.

If voltage is correct, check for 1.3 KΩ (+/- 10%) between fan terminals.
Ed. I have an older fan that has 137 &Ohms;. Runs fine. See HERE.

If resistance is incorrect, replace the fan. If resistance is correct, check spring clip connectors. (If good, there isn't a fan problem).

If voltage is incorrect, unplug fan cable from connector J28 on planar and check cable assembly for continuity. If cable has continuity, replace the system board (uh, unlikely to be bad). If the cable does not have continuity, replace it.

Deriving 17 V DC from the power connector

   OK, you have a 9595-3Px, fully loaded with RAID drives and memory. BUT you notice strange erratic performance after a few minutes. After checking, you notice (to your horror) that you have NO voltage (or not enough) from J28.

   Just trot down to Best Buy and pick up the dual serial/parallel planar for $49.99 and pop it in? Or will you whip out your trusty 25W soldering iron and take charge of your own destiny? If you are of the adventurous type, read on!

> That was my intent - how to derive an alternate source of 17 V DC from the planar power socket...

   Pin 7 of the PSU-connector delivers -12 V, pin 3 +5 V... both add to +17 V. Only need to watch the polarity. (Ed. Verify the polarity with a voltmeter/DMM. Do not assume anything with a 9595 dual serial/parallel planar. One bzzt! and you may burn something else on the planar. The Power Supply will be fine though...)

Tom adds:
   The side fan circuit on the planar is unlikely to fail, as it consists of two direct connections from the main power connector to the fan header. So if you don't measure 17 V DC there, one of the voltages that are used to generate it (+5 V and -12 V) is probably missing. (Note: I have removed the section about the alternative method that used the non-existent -5 V power rail.)


9595 Main Page