“A man with a watch knows what time it is. A man with two watches is never sure.”
I would also say that a man with a voltmeter knows what voltage it is. A man with two voltmeters is never sure.
Between the Solartron 7150, the HP 3456A and the HP 3457A I have three voltmeters… which satisfies Segal’s law.
At least the HP3456A and the HP 3457A read the same and so all is well with the world.
A.Cicuta
Actually you may not need it, but a configure script you need may really really need it.
It happened to me, twice. The second time I decided to write this page as a memo to myself.
You can probably find comm in coreutils. How to install coreutils depends on your favorite distribution.
If your distribution doesn’t package it in coreutils, tough luck, you will have to install it from source.
If you never had to search for comm, I guess you never used it or never run a configure script that needed it.
I also don’t understand why you are reading this webpage then.
A.Cicuta
The Type 602 had been around for a lot of time before I decided to tackle its repair which had been easier than expected.
First problem: missing mains fuse. Replaced and the unit turned on.
Almost everything worked except for vertical deflection of the beam. This also was an easy repair: my unit came without covers and a bit battered so Q27 went missing and Q21 was broken.
They are Tekronix Part Number 151-0195-00 and their equivalent seems to be MPS6515. I did not have any MPS6515 laying around so I went for something similar. In two NPN transistor went and the vertical deflection worked.
I may have to adjust the unit, but I will do that only if I will end up buying the MPS6515.
Anyway, one less broken thing.
The documentation about the Type 602 can be found here.
A.Cicuta
That is the time it took the SGI Octane to do these:
A big thank you to Martin Steen and his frbatch program.
A. Cicuta
This SCI Octane has been waiting for a really long time I could give it the time it deserved but now it happy with its fresh installation of IRIX.
As the machine doesn’t have a CD reader, I had to do a net install. The install server was, obviously, a GNU/Linux machine.
I followed this guide: http://techpubs.spinlocksolutions.com/irix/remote-irix-6.5-installation-from-linux.html with minor modifications here and there.
Now I just need to configure the Octane to install new packages from nekochan.net and play a bit with it. Maybe running some benchmarks or compile something.
I don’t think that this machine will see a lot of use but reviving it has been a lot of fun.
Next on the list machines that deserve time, a Mac motherboard that needs to run again and the a SGI Indigo² that will be installed like the SGI Octane.
A.C.
I have been rather unwell for the past couple of weeks. I’ve had a fever, cough, fatigue… the works. So no updates in this blog and nothing got done outside this blog.
Now I am a bit better but I am pretty busy. While I was lying in bed looking at the ceiling work kept piling up. However, just this blog deserved a little post after the last update.
Moreover things I had ordered trickled in my mail (I’ve been very lucky because I had them not shipped to me… I could not get up from the bed to answer the door).
So now I got my LM399H, li-ion batteries and battery holders, a cheap voltage monitor, a really pretty 5 megapixel camera… and no time to play with them because I have to things to do.
At least the things I am doing are fun.
But I would really like to play with the LM399H. By the way, I measured my REF102B but I have been unable to do any significant long measurement.
Anyway, Happy Easter!
A.Cicuta
I happened to acquire two HP 3488A. They are called Switch/Control unit. That means that each of them can accept up to five different boards that do interesting things. There is a multiplexer, a matrix switch, a digital input/ouput board and so on.
Why would an HP 3488A be useful to me? Because it can be controlled through GPIB and can route signals.
Sadly neither board that I have is a multiplexer. I also do not have the back connector which means that the boards are a bit useless at the moment mostly because I haven’t been able to find the back connector.
So… I am just going to remove the back connector from my matrix switch board and build something. This way I could at least start to use the HP3488A.
Then I am going to remove more connectors and build myself a multiplexer board.
I think I am going to keep myself busy.
A.C.
A new post is long overdue so here it is.
As now I am an happy owner of a nice National Instruments GPIB PCI card, barely enough GPIB cables (you never have enough cables and that’s a sad truth), a Solartron 7150 and an HP 3456A it is time to do some measures and have some fun.
Thanks to gpib-linux the GPIB card works perfectly and I have already connected the Solartron 7150 and started to test it. I already wrote a first, simple script in perl to set the 7150 up and start logging the measurements.
When that script will be ready it will be the HP 3456A turn as that machine is a little bit more complex to set up.
Having the scripts working is a step in the right direction but it will not be enough to just take some measurement. At least I am going to log the temperature at which the measurements are done so I will need a temperature sensor. It will be a good chance to use an MSP430 board but I will have to write some firmware for it. I think that Energia will be more than enough for now.
But what exactly I am going to measure?
Well… I will start with measuring a short then I will go to something more interesting: a little board with a REF102B.
After that, considering that I am waiting for some LM399, the sky is the limit…
A.C.
The 1000V, 100V, 10V and 1V DCV ranges didn’t work right. The Ohms ranges didn’t work either. The self test didn’t pass (error #4).
After reading the service manual I was expecting some problem in the Input Amplifier (a open, shorted or leaky FET). But as the service manual suggested, I first checked the logic.
A CD4555BF wasn’t working right. An output was always high and the end result was that a couple of FETs were always turned ON even when they weren’t supposed to.
The inputs toggled right so I was kind of happy as replacing the CD4555 was less worrisome than replacing even a single FET in the Input Amplifier. It would have been easier too.
Well… I decided to test if the CD4555 was the only problem and so I put the voltmeter in the right range and pulled down the offending output with a resistor, thinking that at worst I would end up blowing the IC output but I was going to replace it anyway.
It worked fine. In fact it worked so fine that the CD4555 started to work in every range when I removed the extra pulldown.
I am going to probably replace it anyway but I am rather curious to understand what happened.
A.C.
… and happy leap second to all the guys who are actually watching it!
I do hope to not miss the next one.
A.C.
