RD54 Stopped Spinning

Jon Elson elson at pico-systems.com
Mon Apr 6 10:41:00 CDT 2015


On 04/06/2015 01:05 AM, Robert Jarratt wrote:
>> -----Original Message-----
>> From: cctech [mailto:cctech-bounces at classiccmp.org] On Behalf Of Jon Elson
>> Sent: 06 April 2015 01:48
>> To: General at classiccmp.org; Discussion at classiccmp.org:On-Topic Posts
>> Subject: Re: RD54 Stopped Spinning
>>
>> On 04/05/2015 04:40 PM, Robert Jarratt wrote:
>>> I tested each FET using a component tester (Peak DCA55) and it
>>> appeared to show that they were all good. I believe they are FETs
>>> because the XT-1140 schematic someone posted shows FETs and the
>>> component tester also reported them as FETs. I do wonder if one of
>>> them is only partially working, I seem to remember that the one you
>>> point out as having smooth ramp was indeed the one with the wobble. If
>>> I have further trouble I will follow Tony's suggestion and swap them
>>> around to see if the wobble moves, but detaching them from the
>>> heat-sink looks to be difficult. Regards Rob
>> I'd just get some FETs that are compatible (probably MUCH better parts
>> available today) and replace any suspect ones with new units. Are these
> pop-
>> riveted to the heat sink?  I vaguely remember a drive built like that.
>>
>> Jon
> I might do what you suggest. They are IRF521 N-Channel and IRF9521 P-Channel
> parts, I'll have to find equivalents. Not knowing too much about the
> significance of the main parameters, I am not sure how critical the
> equivalence needs to be. For example it appears that the IRF521 has
> Vdss=60V, Rds(on)=0.3ohm, Id=5A-8A (temperature dependent), while a
> similarly numbered modern part I found (IRF540Z) has Vdss=100V,
> Rds(on)=0.0265ohm, Id=36A .
>
> The heat sink is not riveted on, but appears to be glued on. I have not
> really tried to remove it yet; to test the parts I removed all 6 of them
> (the heatsink is a single one across all 6 parts).
>
>
There will have to be some electrical insulation as the 
drains of the low-side transistors can't be commoned.

Make sure you select a transistor with equivalent threshold 
voltage (the gate voltage where it starts to turn on).

Jon


More information about the cctalk mailing list