A nightmare on the bench
In its present condition, the 6827A makes a pretty good doorstop.
Brad Thompson, Contributing Technical Editor -- Test & Measurement World, 12/22/2010 12:00:00 AM
|
It has been a bad month for power supplies—my Ethernet router developed bulging electrolytic capacitors, and a newly purchased multi-output "wall wart "didn't work due to unsoldered connections. Unlike the previous items, my newly acquired Hewlett-Packard 6827A bipolar power supply/amplifier is stubbornly refusing to be easily healed. Designed in the early 1970s, the 6827A delivers –10/–100 to +10/+100 DC V at currents up to 0.5 A. In addition, it can serve as a DC-to-15-kHz power amplifier.
Weighing in at 18 lb, in its present condition it also makes a pretty good doorstop.
The 6827A's innards reflect classic 1970s construction: A large motherboard supports sockets for two plug-in boards. An iron-core power transformer rests squarely in the instrument's midsection, and a large finned heat sink carries six TO-3 power transistors. To reach inner components, you remove the cabinet's external panels and displace the power transformer and an attached heat-sink assembly.
|
Unfortunately, the 6827A's designers didn't provide sufficient slack in the wire bundle that connects the power transformer to the motherboard, leaving the heavy transformer and heat sink awkwardly flopping at the bundle's end. Removal of the transformer from the heat sink involves undoing four screws, two of which are blocked by the chassis.
Two ribbon cables respectively connect the rear programming panel and the power transistors' circuit boards to the motherboard. The bad news: Both motherboard connectors are identically keyed. The good news: Both ribbon cables are too short to misconnect. More bad news: They're also difficult to disconnect without skinning one's knuckles.
To date, I haven't unraveled the proximate causes of this instrument's failure(s)—uncontrollable output voltage on its lowest range, no output voltage on its highest range, and a current-fault light under no-load condition. So far, I have replaced dried-out filter capacitors and investigated several overheated components.
So, why rate the 6827A's design as a "nightmare?" I did so because it's impossible to troubleshoot and replace most components and verify the results without disassembling and reassembling the instrument. It's that simple—or complicated.
|
A bunch of gripes An alert reader complained of an
omission in my October column in which I described the DS0201, a
cellphone-sized digital storage oscilloscope that sells for less than
$100; I failed to mention that the scope includes an SD card slot for
waveform storage: www.tmworld.com/article/510727-Take_that_Tektronix_.php |
-
I had a similar problem with a 1970s HP supply 0 to 250V at 200mA.
I found several design boo-boos one was a transistor that would reverse bias the base and break down the emitter-base junction, This works for a while, but over time the beta of the transistor becomes too low, the transistor simply wears out. A simple diode fixed that problem, but now the power-supply does funny things at turn on. I have a slew of other HP design boo-boos, this one has been the most obvious one
Hans J Weedon
Hans J Weedon - 2011-13-1 17:28:35 EST -
We have designed a few new power supply products since 1970. Perhaps you can call 800-829-4444 or visit www.agilent.com and we'll set you up with a new one!
Scott Ferguson - 2011-13-1 16:23:06 EST
No related content found.
- 0 rated items found.
Datasheets.com Electronic Parts & Inventory Search
185 million searchable parts
- Part Number
- Description
- Inventory
- Products
- Manufacturers
View All Webcasts
