> -----Original Message----- > From: cctalk [mailto:cctalk-boun...@classiccmp.org] On Behalf Of Tony Duell > via cctalk > Sent: 28 August 2017 15:52 > To: General Discussion: On-Topic and Off-Topic Posts <cctalk@classiccmp.org> > Subject: Re: PSU Input Rectifier Question > > On Mon, Aug 28, 2017 at 12:49 PM, Rob Jarratt via cctalk > <cctalk@classiccmp.org> wrote: > > My recently acquired DECstation 5000/240 has had a PSU failure. After > > replacing a couple of small capacitors with high ESR, I have been > > doing some basic checks to see which stages are working. My first step > > has been to check the output of the input rectifier, which has two > > large capacitors to smooth the output, they are each rated at 200V. > > When I check the rectifier output with a basic DMM I measure about > > 136V on the AC range, and a voltage ranging around 160-170 on the DC > range. > > > > > > > > I know that some PSUs double the voltage here, and the presence of two > > large capacitors would suggest that happens here too. However, from > > what I have read, the doubling is done on PSUs with a voltage selector > > switch. This PSU does not have an input voltage selection switch. I > > don't know if on auto-ranging PSUs the rectifier output should be 325V > > or 163V. However, given that each capacitor is rated at 200V then it > > seems to me that the rectifier output is expected to be 325V. Is that right? > > If you havr 230V mains input and you full-wave rectify it and smooth it (say > using a bridge rectifier) you will get 230*sqrt(2) or about 325V. There is no > easy way to get 163V DC from a 230V AC input (you could do it with a step- > down transformer or a voltage regulator circuit, of course, but not with just > diodes and capacitors). > > The common input circuit for PSUs with a voltage selector switch is quite > ingenious. With the switch open, it's a bridge rectifier and smoothing > capacitor > (the latter is actually 2 capacitors in series, with bleeder resistors, each > charged to around 163V). With switch closed, it becomes a voltage doubler. > Two of the diodes do nothing, but again each capacitor charges to around 163V > (with a 115V AC input). > > A lot of older auto-sensing power supplies used that circuit, with an > automatic > switching component (triac or similar) in place of the switch. > > The first thing to check is if the 2 capacitors are in series. Does the +ve > terminal of one go to the -ve terminal of the other? What voltage do you > measure across each capacitor separately? > > What is your mains voltage? 115V or 230V? If the former then it's possible the > automatic switching has failed and that it's acting as if it is connected to > 230V > mains. >
I have posted some information along with a rough schematic here: https://rjarratt.wordpress.com/2017/08/28/decstation-5000-model-240-h7878-possible-fault-found/. Other information is in the previous post: https://robs-old-computers.com/2017/08/27/decstation-5000-model-240-h7878-power-supply-failure/. Interesting that Tony mentions a Triac. There is indeed a triac. I am using 220/240V input (being in the UK), would it be expected to double the voltage in this case to 325V? Regards Rob
