Seed thread to start things off.
|Peter G. Shaw||02/06/2014 16:14:56|
1191 forum posts
I wonder why all the valves as used in my very first amplifier are now readily available again. Could it be because the sound is perceived to be better that semiconductors?
Like a lot of other stuff, the 4000 series has been around for something like 40 years which to me says something about its usefulness. As to how long will it be around, who knows. As I said some valves are still available.
And of course, that's why home maintenance of cars is falling by the wayside. As is home maintenance of almost everything else.
Of course things will change in the future, but not necessarily for the better. In my opinion, we are rapidly heading towards a situation where the man in the street will not be able to do anything for himself: instead anything that goes wrong, and they will regardless of how much improved they are, will require the attention of a highly trained, and expensive, person to sort it out. We already have people who can't change a plug for example. Or change a tapwasher. Or sew a button on (I'm married to one).
Inappropriate applications. And that brings it home. Do it this way, or it won't work properly and might fail. Where's the creativity there for the amateur - there is none. Where's the incentive for the amateur to learn anything new if all you can use are little black boxes. It's a similar argument in respect of CNC against manually controlled equipment. I get enjoyment and increase my skill level by doing it manually whereas programming a CNC machine does nothing for me at all as once you've satisfactorily programmed it, what then? At least with discrete components I can learn and adjust to suit my requirements. I've nothing against programming by the way: I keep looking at a project at the moment, but realistically, it won't happen because of time constraints, both in the immediate future and lack of future years. But nevertheless, thinking about it does help to keep the braincells ticking over.
If you want to use black boxes and simply make it work without understanding what's happening inside, then fair enough, you do that. For me, that isn't good enough as I want to know and understand the totality of what I'm doing. And by the way, this doesn't mean that I need to know all the ins and outs of my computer, my television, my Freeview adaptor, my ...... etc. I am referring purely and simply to anything I might wish to do as an amateur.
Regards, and bye-bye,
Peter G. Shaw
|Neil Wyatt||02/06/2014 21:58:20|
18320 forum posts
I hope you're not waving goodbye for good Peter?
I'm afraid I have to take a different view on some of those issues. Valves sound great. They are useless for accurately reproducing sound at volume, but they sound nice when they distort.
I have one of the new breed of guitar amplifiers. it uses a small valve amp with a microprocessor to simulate 33 different guitar amplifiers. The microprocessor is damned clever and can imitate anything from a Fender Twin, and AC30 to a Marshall stack - but it still needs to overdrive a real valve to get the authentic sound. ironically, the power amp stage is solid state - so that it reproduces valve-distorted sound correctly at any volume.
So the latest and the oldest technologies live together side by side in harmony (or infernal racket, depending which amp model you choose). Why can't they do so in our hobby, especially as this isn't 'electronics for electronics sake'. This topic is 'electronics as an aid to the workshop', so the right choice for anyone is what makes their hobby more satisfying, and only each individual can decide that.
I knew someone who claimed my 'computer' I made wasn't a real one as I should have programmed the processor into an FPGA. Well I have come across a guy who made a microprocessor out of TTL and wire wrap...
But then you could argue that Acorn didn't make the BBC computer because they used Rockwell's 6502 instead of their own processor, or I didn't 'make' my 10V because I bought the castings from Stuarts, or Norden, because someone else poured the castings from my patterns..
It's all a matter of degree and personal satisfaction. It's also means to an end. Microprocessors let you make a lot of things you just couldn't, practically, do without them, but as I said earlier, you need to understand 'discretes' if you want your black boxes to talk to the outside world.
Jack Spratt would eat no fat, his wife would eat no lean, yet between the both of them, they licked the platter clean.
|Peter G. Shaw||03/06/2014 13:27:33|
1191 forum posts
No I'm not leaving for good, just not taking an active part anymore in this particular thread.
It's obvious that my ideas on simplicity, ease and understanding for amateurs are not well received so there's nothing more to be said.
Peter G. Shaw
|Andrew Johnston||07/06/2014 12:17:44|
5726 forum posts
I have experimented with one of the processors embedded within an FPGA; just to see if it worked, I never used it in a real design. I think it is probably a solution looking for a problem. A basic processor is fine, but not very useful. Gone are the days of adding support chips; now everything is expected to be on the processor IC.
There seem to be some mis-conceptions within this thread regarding professional versus amateur. So here are a few thoughts on the matter.
I have used high end design tools (£100k+) in the past, often at the behest of the client. However, for my current designs I use Geomagic (Alibre as was) for 3D CAD and Easy-PC for schematic capture and PCB layout. Not much different to the keen amateur I suspect. I do all my own PCB layout. That's one big change that has happened over the years; the glory days of external PCB bureaus are over. It's part of a trend, you don't see many typing pools these days.
I don't have access to much more information than the amateur. I use datasheets, apps notes and the like from the web. I do have contact with suppliers and reps, but that's more to do with component availability and costings than technical. There are online help forums, but they generally reply with a pointer to the datasheet - useless, I've read the datasheet, and the answer isn't there, which is why I asked the question!
In some cases you have to register to get data; I think the last time I had to do that was for DDR SDRAM, not something I suspect the amatuer is likely to be using. There are also ICs for which NDAs are required even to get a datasheet, but these tend to be highly specialised RF and consumer type devices, which very few amateurs, or professionals, would be using. There's a whole range of semiconductor companies out there that exist to supply the high volume consumer electronics markets, and rarely sell via normal distribution. In the past I've been told that they were not interested because I was 'only' asking about 100,000 parts per year.
Mention has been made of designing with black boxes. Modern electronics is like that, analogue or digital. There is no way one is going to understand the detailed internal design of even a fairly simple microcontroller. But then again how many people really understand the internals of an opamp? And no it doesn't look like a conventional amplifier circuit with transistors, resistors and capacitors. Think more in terms of current sources, current mirrors, active loads and circuits that rely on transistor parameters that are set by the physical size of the device.
One big difference between amateur and professional is that the professional has to deal with a number of 'soft' factors that the amateur can ignore. These include things like cost, ease of assembly, component supply chains, performance across temperature, EMC and ESD, the list seems never ending!
One last thought before I head back to the workshop (all for work unfortunately, the traction engines sit neglected in the kitchen) is the question of creativity. To be honest there isn't that much scope for creativity in general electronics. The really exciting areas are in IC design, where it is still possbile to be blown away by a design features or a new approach.
Here is an example; I recently had a requirement for a power supply to generate 0 to 10V at 500mA, and to be controlled by a microcontroller via a DAC. The classic feedback loop in a power supply uses a potential divider to sample the output voltage and compare it with a fixed reference. So far so good, we just need to 'interfere' with the voltage on the potential divider to control the output. But there are two problems. One the output voltage is unlikely to be a simple function of the control voltage, and two what happens when the output voltage reaches the reference voltage? The answer is that the minimum output voltage is set at the reference voltage. After trawling the internet I found a Linear Tech switcher IC where the reference voltage has been replaced with a 50µA current source. Drive that current through a resistor and the IC will drive the output to match the resulting voltage, possibly through a potential divider. But there is no limitation on the voltage, if we choose a resistor that only generates 100mV, the output can be set at 100mV. Now, suppose we drive the current source pin from an opamp that will sink at least 50µA, the output will follow whatever voltage is at the output of the opamp, right down to zero. Really, really neat. The output versus control voltage is linear, and the output will go to zero. Even better, although not relevant in my application, is the opamp output doesn't have to be DC. I bought the devkit and adapted it to my design to check out the performance, the output voltage into a 10 ohm load will follow the control voltage up to several hundred Hertz.
Good grief, that was a bit longer than anticipated, I expect everybody is now asleep - Zzzzzzzzzzzzz.
DDR SDRAM = double data rate static dynamic random access memory
NDA = non-disclosure agreement, usually between companies to protect IP
IP = intellectual property
DAC = digital to analog converter
|jason udall||07/06/2014 13:51:34|
|2028 forum posts|
|Mmmm...andrew...Powerop amp maybe...I made myself ( back when you sometimes needed to build test gear)..a dual channel 50 -0 -50 5 amp"programmable" power supply.. based on a power op amp... .nice chip..10 MHz gainbandwidth product...fine for me only asking gain of five and bw of 100 K....|
I also remember the 700 series op amps..even has to do a lab at uni ob them making a replica in discreet transitors......
Ahh the nostalgia. .
But then again portability and design cycle is the thing...so code for fpga..is where it's at...most of the time...
There are still times requiring...finesse. ..
But for one off's ..well chuck a micro controller at it and get it done in code
|Neil Wyatt||07/06/2014 15:24:02|
18320 forum posts
What you end up with sounds suspiciously like a Class-D audio amp...
As for the cost of getting into 'black boxes' Arduino Uno is £49+VAT bundled with breadboard and enough components for anyone to get started, from CPC.
|Les Jones 1||07/06/2014 16:46:13|
|2171 forum posts|
The TI MSP430 Launchpad is less than 10 pounds but the development software is much less intuitive than PIC or Atmel software. I bought one of these but have not yet done much with it.
|Steve Withnell||07/06/2014 19:47:51|
821 forum posts
I've just picked up another transceiver to restore. The thing is that choked with nicotine/cigarette tar you need an NBG suit on before removing the covers. How the hell to you get rid of all that crap on the pcb's and interstage wiring? The radio looks like it's had a 60 a day senior service habit for the last forty years.
|Billy Mills||07/06/2014 22:18:58|
|377 forum posts|
You need to wash it. Electrolube used to make a very good water based flux remover in aerosol cans called Safewash which had an attatched brush. We used to clean up pcb and monitor chassis/pcb's without issues although LOPT's were not cleaned because the focus assembly is not hermetic. Providing that you can compleatly rinse and dry -if you know what you are doing- you can wash clean almost any electronic equipment that is water safe. Washing should remove trace lubes on rotary switches and mechanical parts so remove or relube.
For cleaning at home warm water and washing up detergent works well, Would not want to wash any LCD displays or non sealed relays however, too many layers, nooks and cranies to drain.
Back in the days of rental tv some service departments used to dunk the insides of tv's in ultrasonic cleaners that used an innert liquid. The liquid was demonstrated on Tomorrow's World by lowering a working TV into the liquid. Most U/S cleaners were run with water and detergents for equipment cleaning, very much cheaper.
Was amused to read about valve amplifiers and their wonderful sound!!!!! well back in the day... I had holiday jobs in the electro-music industry when I was a kid. Used to fix Hammonds, Vox, Marshall, Fender, CopyCat et al.... The main design principle used in most equipment was that you could not make it any cheaper. They got away with it because amplifiers ony ever worked on one signal source so the intermodulation, limited bandwidth and resonances were not apparent, only part of the "instrument" sound. Made a bit of money cutting Hammonds in half- called "splitting" and adding pedal sustain and other effects.
|Geoff Theasby||08/06/2014 12:54:31|
|613 forum posts|
In the matter of solving simple but clunky problems by throwing technology at them, the current Practical Wireless has a news item on proposals to control the peripherals of a motor vehicle using Bluetooth equipment, incorporating specially designed electronics capable of coping with the harsh environment in such a situation. Much less copper wire and complex harnesses or looms, no relays, no individual switches, but presumably supplying the current through a single bus wire.
|Neil Wyatt||08/06/2014 15:53:14|
18320 forum posts
I came across a professional electronic engineer who used to wash boards in the dishwasher.
|Neil Wyatt||08/06/2014 15:54:50|
18320 forum posts
> control the peripherals of a motor vehicle using Bluetooth equipment,
Authentication required. If you wish to continue braking, please re-enter your PIN.
|Danny M2Z||08/06/2014 16:34:37|
892 forum posts
I remember when the U.S. lost a bunch of the first Blackhawk helicopters in Europe. (At $6M a pop)
Turned out that the 'fly-by-wire' systems picked up the Megawatt signals pumped out towards the east by Radio Free Europe, powerful enough to inductively develop dc voltages in the servo control circuits.
I hope the Bluetooth is only going to control the back seat movies.
* Danny M *
|Billy Mills||08/06/2014 17:50:30|
|377 forum posts|
There was an Aircrash Investigation show on the Air France Airbus into the Atlantic on a few days ago. This was a very carefully designed aircraft and the crew were especially trained to fly the aircraft using sidestick controllers.
They flew through some weather that iced the pitot tubes, one of the pilots misread the instruments and pulled the sidestick hard back. The aircraft stalled at 33,000 ft. The co pilot then applied nose down on his side but the first pilot still held the stick hard back which negated the second stick input. The aircraft fell into the sea, everyone died.
Had the controllers been mechanical or had some feedback it would have been obvious what the first pilot had done, the incident would not have happened.
So Bluetooth and car designers?
2904 forum posts
That Airbus story as recalled seems to have been passed through the "Daily Mail Science Correspondent" filter. I'm not an aviation expert but even so I recall there were a few more details to the plot. I seem to recall they were both disoriented and weren't even aware the plane was in danger of stalling until too late. A mechanical coupling wouldn't have made the slightest difference. IIRC there had already been reports of problems with the Pitot tubes icing but nothing was done in time
As for cars being somehow controlled over Bluetooth and going doolally, I don't think you would believe how many layers of safety and redundancy there are in modern cars. Apart from the schadenfreude from the US car companies over the supposed problems with Toyota's throttle pedals (which turned out to be mostly false reporting and hysteria), I don't recall hearing of lots of issues with robustness on the engine control front. Cars nowadays are far less likely to suffer from loss of control than in the good old days (sticking throttle cable, anyone?) and nobody is about to throw all those babies out with the bathwater. Makes for an easy chortle down at the pub but quite simply there is no likelihood of that ever happening.
|Andrew Johnston||09/06/2014 10:41:15|
5726 forum posts
Good point, the output of a class D amplifier is akin to a buck converter with synchronous rectification. However I suspect that there are some subtle differences in the control and feedback circuits. Despite the novelty of being able to amplitude modulate the PSU output, in my application the important factors are DC accuracy and stability; possibly not so critical in an audio amplifier.
When I was talking about black boxes, I was referring to ICs rather than devkits, and was commenting that very few people really understand what goes on inside the IC, even simple ones. Certainly within the professional world the devkit has assumed new importance. A combination of complex processors and short timescales means it is not possible to understand everything in detail, so a devkit is very useful for getting ideas on how the darn thing works, and how to make external connections. Modern processors have datasheets and reference manuals ranging from hundreds to thousands of pages. That's a lot of reading, and even then some things are not clear. The main bugbears are reset procedures, power supplies and sequencing, and multi-function I/O pins.
|Cornish Jack||09/06/2014 11:23:13|
|1184 forum posts|
Murray - Hammer, nail, head
2904 forum posts
It's often tempting to believe that you can create a robust design simply by stringing together black boxes and devkits / eval boards. However, if you have no real, profound understanding of what the black boxes are doing, you are sort of hoping they do what you expect them to do. That would be nice. Unfortunately, there are often several ways for a manufacturer to achieve a given "black box" functionality and it's often essential for you to put the time in to figure out what's happening if you are to avoid those sticky moments.
As Andrew says, you can't possibly get right to the bottom of what's happening in ICs but as he mentioned there are often a few critical details hidden away in the app notes that will bite you if you don't pick up on them - and even a few that aren't! You soon learn that it's less time consuming in the end to do that homework up front than to rush the first stages and deal with the issues later. To my mind, the cost of rectifying underlying issues increases something like logarithmically, the later in the program you leave them.
|jason udall||10/06/2014 18:11:54|
|2028 forum posts|
|As to not knowing whar goes on inside the black boxes...I remember writing custom assembler to exploit undocumented op codes for both motorola and intel processors...and enjoying the re appearance of some of these (intel 808x) codes in the zilog z80.. ( explain that one)..|
Best mnemonic ?...HBF..Hault and burst into flames..which is pretty much what it did. .
But yes you can't have a micro code knowledge of every chips innards but design from data sheet should work.......
|1038 forum posts|
I amused my wife this past weekend, I had been asked to repair a laptop which had been treated to a large glass of sherry. The smell of sweet sherry was quite overwhelming and suffice to say, the laptops behaviour was somewhat erratic.
So, complete strip down, bare case in the sink with hot soapy water, along with the keyboard. The main PCB was treated to a liberal dose of a product called safeclene, rinsed with clean water and all components hung on the washing line in the sun to dry. (not sure what my neighbours thought )
Re-assembled last night and fingers crossed, applied power, SUCCESS.
I've done a few like this in the past, mostly kids throwing their can of coke over it, this was the first alcoholic laptop I've had to deal with.
Please login to post a reply.
Want the latest issue of Model Engineer or Model Engineers' Workshop? Use our magazine locator links to find your nearest stockist!
You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.
Click THIS LINK for full contact details.
For subscription issues please see THIS LINK.