If I had to pick a plan out of the air this week, I’d try the Amazon EC2 micro instance of windows server. It’s free, and you’ll learn a lot.

But it’s a moving target. I strongly recommend relying on ‘the cloud’, because you can be instantly locked out of your site for no reason. And it’s happened to me.

I would also like to see a future post that shows the cost and payback of the system over time. Nice to see these systems able to grid tie so that batteries are no longer needed.

I am surprised at how different your electrical panel look compaired to north american ones.

There is a large windfarm in texas, you can pretty much buy credits at it and say you’re “wind powered”

This guy is the first thing that came to mind when I saw that you were giving these guys a recommendation… Hilarious

He’s getting all the tattoos removed by one of the companies now so it’s got a kinda happy ending

Spot on man…. My company just got bought out and I saw 20 years worth of custom software development for our specific industry tossed in favor of their archaic piece of crap system. We went from a lean agile organization that got things done to one that had to have meeting upon meeting run by people who didn’t even understand the problem. The best I can say is as of last Friday, I’ve got a great severance package.

I thoroughly enjoy your blog. Keep it coming man.

Paul

Nice work as usual – but a quick question:

Can you predict the limits of output current using the reactance of the caps (i.e. Xc=1/jwC)- so you can design a supply in rather than measure it?

Thanks,

Simon

I think it’d worth a post of its own.

With regard to #484 I think I got a glimpse of a potential new blog.

DBVblog: “Dave’s Botanical Video blog”. And, to quote you: “Ahh. . . man, I don’t know, maybe,”.
Cheers,
Mark
******************

I also can’t believe that you said that you bought a vacuum cleaner for your wife. If you said that in the UK you’d be the one getting a rectal insertion.

This multimeter probably cost a lot of money back then, at least several hundred dollars today, if not thousands. Sure it’ll last decades, because people expected it to for the price.

It’s just like everyone complaining TVs and such are crappy – well, that decades old one could easily cost over a year’s worth of disposable income, while today you can get them for a month’s worth of disposable income. Of course, you also were able to repair them because you couldn’t go out and buy a replacement without saving up for it. When they cost lots, repair was worth it – now just doing a diagnostic will cost more than its worth fixed (and it’s only good if you don’t value your time and can spend hours fixing it – for fun or a hobby).

You can find quality stuff these days still, but like years gone by, you’ll have to ante up for it. It’s just that these days, instead of heavily overengineering the stuff, people decided making it more accessible by making it cheaper was better.

And PICs with flash have always had 1000 write/erase cycles – I saw them going way back. It’s mostly because of the processes used – usually the big guys like Samsung etc., have special lines for making flash memory (and DRAM), and other lines for generic CMOS logic. While they’re similar, the specialized line is able to make better quality transistors to give better endurance and capability. The generic line can manufacture them as well, but not as good. Microchip and others do the same thing – you work with the existing process, and the quality generally isn’t as good as it would be if you used a separate chip.

Look at this high impedance op-amp app. note from TI. Pay particular attention to electrical guarding for the op-amp inputs. It may help you in your case.
http://www.ti.com/lit/an/snoa664/snoa664.pdf

21W is standard rating for the front and rear bulbs, 4W for the side repeaters or a warning light.

Bulb resistance is used in the circuit to vary the flash rate which indicates a bulb failure by flashing about twice as fast. Legally required audible warning is provided by the mechanical clicking of the relay armature against the pole piece, otherwise a fully solid state solution would probably be cheaper, but not isf an additional sounder is required. It also sounds the same as the old thermal units so no surprises for the uninitiated.

In fact is a shame that today engineers are paid to develop devices and products with reduced durability (as the last series of PIC32 and PIC24 with no more than 1000 erase/write cycles and with a lot of silicon bugs). And we are preparing to conquer the Mars…

#4) Yes, I goofed this. The ideal effect of Ib x RF does not get multiplied by the gain, it is the direct (ideal) output offset due to a single Ib.

So there is collection of issues at play here. But the end result shows that I was able to get the final output offset voltage to within expected limits by compensating for Ib and reducing the Rf ratio.

try to salvage those power supplies from the HP server and try to make a nice +-12VDC and 5VDC power supply of it for homebrew.
Maybee you can even hook up a solder iron.

Sorry for my bad English.

Keep up the good work!

Have you explored the OPA2277 chip yet?

These are machines too expensive to get us locked at the hands of a manufacturer. It is particularly frustating to me, as I do sw for a living and can design/code at any level. I would love/needed to replace my DS1052 for a faster one, but I just can’t live with another noisy fan’d scope nor with an ugly, buggy and un-ergonomic interface nor with a very expensive and feature lacking one. It’s scope dead-end for me right now. Would it be too hard to develop a 2/4 channel 100MHz 500/1000Msps open hw scope?

the coloured strips you took out of the box at minute 27 and want to give to Sagan are very useful for marking pages in books. I used a lot of them in my studies because they are very useful if you work with library books.

Keep up the good work!