nt fails. If 50 year old valve kit is working it has proven itself, and any new electronics has a very low chance of lasting that well. So in practice old valve kit is on average more likely to last better.
es not prove that it is a reliable gear.
It proves the design is capable of great longevity
ept for one.
There's always a 1 in 50k chance of that. Statistical conclusions are only a reflection of the data fed into them.
1500 years.
And the mean time to failure for soldered connections, electrolytics, bits of plastic, switch contacts and all other parts?
s of plastic, switch contacts and all other parts?
On the stuff I was designing there were no bits of plastic, switches , or soldered connections. The electrolytics were the biggest problem and we u sed mostly dry slug tantalum caps ( as I remember it, It was a while back )
It was somewhat after someone at MIT or Harvard built a difference analyser with something like 50,000 tubes. The tubes had a 50,000 hour average lif e . Se it would fail about every hour. To find the bad tube took somethin g like two hours It was not a success.
I haven't read all of the posts in this thread, but I see a lot of confusion of the terms "function generator" and "audio generator". Audio simply refers to the frequency range of the signal generated without saying anything about the manner in which it is generated. But most signal generators output a sine wave and perhaps a square and triangle wave. As others have said, they were often analog circuits that shaped one waveform into another. This is different from an "arbitrary waveform" generator which some seem to be calling a function generator.
An arbitrary waveform generator has a pattern in memory which is played through a DAC to produce any function you wish which must repeat only when the memory capacity is reached.
A DDS is another type of repeating waveform signal generator that also uses digital logic and a DAC. It uses an incrementer to model the phase of a signal cycle which then is translated into a waveform and played through a DAC. If you want a sawtooth wave, you just play the phase value through the DAC. A sine wave may be generated using a lookup table or some simple sine generation functions. Other waveforms can be generated using the lookup table or other mathematical functions. But all these signals must repeat at some period as defined by the amount the phase register is incremented at each clock. In fact, that is the strength of the DDS over an analog signal generator, the frequency of the output signal is very accurately defined.
I'm not sure why you think these devices would not be reliable or are inherently inferior. The Ham community have been using DDS for some time now. Even if they aren't used to directly generate the frequency needed, they can be used as a tunable reference frequency with a PLL to generate a very accurate frequency of nearly any rational value.
its of plastic, switch contacts and all other parts?
or soldered connections. The electrolytics were the biggest problem and we used mostly dry slug tantalum caps ( as I remember it, It was a while ba ck )
Most testgear of course is built to nothing like those standards.
er with something like 50,000 tubes. The tubes had a 50,000 hour average l ife . Se it would fail about every hour. To find the bad tube took someth ing like two hours It was not a success.
A killer problem with valve computers. At least they were better than relay s. A sig gen with a few valves is a very different proposition. Basic maint enance every 16k hours of use, typically several decades, is quite acceptab le.
Wikipedia often is written by those with a certain level of elitism rather than an interest in explaining topics to all who wish to learn. It is not uncommon for a technical article to be written at such a high level that a reader needs much more than just a casual understanding of the topic.
In my opinion, this is one of the ways Wikipedia has failed.
How exactly would a computer distract you while you are soldering??? That is very unclear.
If you want to deal with tube circuits you need a buffer between any low voltage equipment and the tube equipment.
Maybe there's a reason why tube equipment isn't made much anymore?
Try connecting with a Ham club. They often use old equipment.
The problem with buying wall warts is getting one that does what it says it does. If you type in 5v, 1A for a wall wart on ebay, the first hundred or so units have the same picture, are priced within a nickle of each other and contain a single transistor for the entire active circuit!
You could just insert a tube amp between the DDS and the gear you are testing...
It's just a single board computer with plug in modules for a wide range of I/O functions. It also has one of the easiest to learn development systems for writing code. If your DDS has a serial port for control the Arduino could be used to make it do more complex functions by changing it from this waveform to that waveform in real time.
That's incorrect; most inexpensive signal generators output a square and triangle (hey, it only takes two op amps); adding a sinewave is complicated. To do a GOOD sinewave, the old HP20x units had matched pairs of adjustable capacitors, which are VERY pricey items.
So, the iCL8038 and XR2206 and some other IC generators (which dominate the market at the low-cost end) distort and/or filter the triangle wave to make a "sinewave". The sinewave outputs are dreadfull.
For audio test purposes, a CD with test tracks is a pretty good sinewave source. Digital sythesizers are good, too. Neither is convenient for a test bench, though.
** It proves that someone has been keeping it running.
** Which only means that the unforced failure rate is one in 1500 per year. Says nothing about the typical lifespan and it sure as HELL ain't 1500 years.
Weeks? MTBF of 336 hours? You can't aford a tube's filament current? GMAFB I guess you never turn on a light bulb, either. Heaven forbid that you would make diner!
I'm not clear on what you consider inconvenient, but you don't need much space to create a decent sine wave. I produce a board with a stereo CODEC and during test it produces a sine wave that I can't visually detect and defects in at any resolution. If I removed the application specific items and gave it a useful PC interface it could sell for maybe $50. It would have perhaps $20 worth of parts and could fit in an Altoids tin easily. I believe there are similar products on eBay which sell for less.
Not a sinewave suitable for a set of test results that can be easily analyzed. The "sinewaves" have all the harmonics of a triangle wave, but at lower amplitude, and this is accomplished with hand-tweak adjustments.
If/as you want to test an amplifier for distortion, that 'sinewave' is not pure enough to use.
What is inconvenient, is that one cannot dial the frequency with an analog knob, while listening for room resonances or speaker-crossover blips. A DDS solution will have to be programmed to do a sweep, and reprogrammed if you want to check that one part of the range that sounded odd... or at least, that was how the old Agilent DDS worked.
nt. So you are better off buying something new. Getting a replacement vacu um tube may be nearly impossible.
When I was in the Navy I had a mk 37 gun fire control system to maintain. The radar used a fair number of vacuum tubes. When the 5 inch 38 guns were fired the radar would go down after about two salvos because some tube ha d failed. So the system had a MTBF of about 5 minutes or less.
It was never anything like as good as a Wein bridge sine wave but it was good for about 0.5% THD if you trimmed it properly. I suspect manufacturing tolerances made it inconsistent batch to batch.
Cute chip in its day, but that was a long time ago.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.