Sig-Gen/Function Generator/Audio Oscillator Purchase

[allen.gordon]

I have researched my questions throughout the forum and encountered some helpful information, but also some confusing information. The culmination of this search and my own desires leads me to the following questions.

First, here's some background. I attended a basic electronics course focusing on guitar/bass tube amplifiers, including effects pedals. I currently repair tube amps using my Fluke 83 V DMM, schematics, and a calculator. I'm able to troubleshoot/repair, but it costs me a great deal of time! I'm losing money on amp repairs without proper equipment.

A 'scope has been selected and will be en route to me soon. FYI, I selected the Rigol DS1054Z as recommended. I will purchase it from www.tequipment.net when the purchase order has been approved.

I will expand my interests to include high-fidelity tube amplifiers in the future.

Money is a concern, but I need the most accurate sinusoidal waveform with the least amount of distortion at the best price. If it costs a little more than an entry-level piece of equipment, I'm willing to pay for those specifications.

I also WELCOME any suggestions or even questions concerning my use of the proper generator.

1) Which would suit my needs best? (Injecting an accurate low-distortion sine wave into audio gear)
2) Can you list a specific make/model so I can research prices.

If I skipped the release of any pertinent information, please inquire.

Thank you, in advance, for your assistance!!

Allen

[Lightages]

For audio frequencies it is hard to beat your sound card on your computer. It is free and very low distortion. You can use a good free program like this:

http://www.sillanumsoft.org/

[allen.gordon]

Thank you! I've downloaded and am using this software now.

I will need a stand-alone unit for the shop, though.

[Andy Watson]

For audio frequencies it is hard to beat your sound card on your computer.

A good tip, but bear in mind that accidentally touching the wrong wire in a valve amp can easily send several hundred volts up the output of your sig-gen - it might be wise to put some protection in line and/or choose a sig-gen with an easily repairable output stage.

You can use a god free program like this:

I always like my programs god free

[f5r5e5d]

a electronics shop without a PC???

DAP - Sansa Clip? you can make audio test .wav, sines, sweeps

knobs, dials are handy but it really hard to beat digital audio sources for audio signals

[deadlylover]

Yeah, using a PC with a sound card or something like the QuantAsylum QA400 is a good option.
That might be the cheapest/easiest way to go if you have a PC in your lab and you don't mind fiddling with it in your workflow, you'll get performance on the order of THD=0.0012%/THD+N=0.0014% @ 1kHz (0dBV 20kHz BW, loopback through it's analyser).
Unfortunately, it doesn't specify the output from the oscillator only.

If you need a step up from that, you have a couple more options.

The Panasonic VP-7722A audio analyser (which I own) has an excellent audio oscillator built into it. It performs at THD=0.00007% / THD+N=0.00026% (0dBV 1kHz @ 30kHz BW). Now, that's a loopback including it's analyser section, so the oscillator must be damn good. (the spec sheet isn't quite clear what it's oscillator is specced at)
If you're patient, you could find one for about $600 delivered. It's an excellent instrument, I doubt any tube amplifier out there can outperform it. =P
I really like it because I don't have to dick around with a PC to get results, it can measure separately down to the 5th harmonic at the push of a button, too easy.

Finally, for an oscillator only, the ShibaSoku AG15 is rare as hell but it is a good performer, from 100Hz-10kHz it's typical THD+N<0.000125%.

If you only need a few spot frequencies, you can probably DIY an oscillator fairly cheaply.

[dom0]

There are literally tons of used audio oscillators out there made by hundreds of companies. For most tube amps they will probably easy outperform the tube amp, even if it's from the 60s. Almost all of them can be cheaply upgraded to excellent performance for a few bucks, as the main cost of a very good audio oscillator is
- case
- power supply
- switches
- lots of rheostats or switched resistors
- output attenuator(s)

In contrast to the above half a dozen of NE5532 cost just a few dollars and the other passives required are cheap and somewhat few in number.

Notebook sound cards usually have good low distortion performance, but often suffer from high noise content (also true of most on-board sound cards in general), which makes them basically useless for generating low-level signals. They also won't give high level signals beyond typically ~1.5 Vrms. They usually have no DC offset control, which is handy, and tend to have poor frequency response (obviously) below 20 Hz or beyond 20 kHz (which makes them useless for testing frequency response, since most audio amplifiers go beyond that.). Some units suffer of hefty jitter issues which manifest as sidebands. Some units have very poor to essentially useless SFDR at low signal levels (idle tones).

But for a start they're fine. Consider some output protection, though.

[allen.gordon]

a electronics shop without a PC???

I have a rather nice tower I built specifically for recording audio demos for select clients. I don't actually use my discrete soundcard, as I use FireWire connected audio interface and near-field reference monitors. The VA-software is useful for research and development, but not so much for modifications/repairs because it's not directly at my workstation. I'm trying to maximize my time in the "Money Square" which includes limiting the time spent online.

That's not to say I will not benefit from the PC-based application. I have spent some time familiarizing myself with the controls/functions of the VA program.

The Panasonic VP-7722A audio analyser (which I own) has an excellent audio oscillator built into it. It performs at THD=0.00007% / THD+N=0.00026% (0dBV 1kHz @ 30kHz BW). Now, that's a loopback including it's analyser section, so the oscillator must be damn good. (the spec sheet isn't quite clear what it's oscillator is specced at)

THAT is seriously low THD! I will definitely check into the Panasonic VP-7722A!!! Thanks for the suggestion!!!

[Kleinstein]

A sound-card offers both a low THD source and the option to analyze the signal. However you need some Output and input protection and likely amplification / attenuation.

[dom0]

THAT is seriously low THD! I will definitely check into the Panasonic VP-7722A!!! Thanks for the suggestion!!!

Most other audio analyzers have specs in the ballpark ; but some are more expensive than the Pana ones, even if their specs are worse (8903x, for example).

Also, I personally have a strong dislike for giving specs like THD, which is a level and thus inherently perceived on a log-like scale, on a linear scale (e.g. percent). I think many people do that just so they can bask in the glow of many zeroes behind the decimal point. 0.0001 % ~ -120 dB

[tautech]

I also WELCOME any suggestions or even questions concerning my use of the proper generator.

Check if the Siglent SDG series meet your needs.
http://www.siglentamerica.com/pd.aspx?tid=16&T=1

Thread for the 1000 and 800 range of models:
https://www.eevblog.com/forum/testgear/the-sdg1000-and-sdg800-thread/

[edavid]

An HP 204C, 204D, or 209A is a good, cheap audio oscillator.  The distortion is around 0.1%, better than any tube amp.
You should be able to buy one for $20-30.

[allen.gordon]

I have been searching for an HP 209A ALL day... I found one listed on eBay, but by the time I signed in to purchase it, someone had grabbed it only minutes into the listing. It was for $90 As Is - No testing... PLUS shipping!

I have found one for sale at a testing equipment site, but it is $275!

I understand that tube amps have high THD by-design, but that doesn't mean I'm going to inject a signal with equally high THD to perform any testing.

I have found rather old threads on some generators, but they suggest units I'm unable to find readily available.

[edavid]

I have been searching for an HP 209A ALL day... I found one listed on eBay, but by the time I signed in to purchase it, someone had grabbed it only minutes into the listing. It was for $90 As Is - No testing... PLUS shipping!

Definitely not worth that much.

I understand that tube amps have high THD by-design, but that doesn't mean I'm going to inject a signal with equally high THD to perform any testing.

It is not "equally high", it is 10X better, which is good enough.

You could also look for a Leader LAG series generator.

[allen.gordon]

Thanks to all for your input and feedback! It is greatly appreciated!

Unless someone has some substantial information not available in the many threads, I have decided on the Siglent SDG1010. It is overkill for my intended use at the moment, but I'd rather have it, than need it!!! 

[Lightages]

The SDG805 supposedly has lower distortion.

[tautech]

Thanks to all for your input and feedback! It is greatly appreciated!

Unless someone has some substantial information not available in the many threads, I have decided on the Siglent SDG1010. It is overkill for my intended use at the moment, but I'd rather have it, than need it!!! 

I'd have thought you might have seen this post:
https://www.eevblog.com/forum/testgear/the-sdg1000-and-sdg800-thread/msg644687/#msg644687

It shows the SDG800 series as having much better THD stats.
But if you want 2 channels.......

[allen.gordon]

 

I was in the first few pages of this very thread when I posted my decision. In fact, I had the thread pulled up in another tab and didn't realize I had been sidetracked by the cross-references in that thread.  I just closed 54 out of 56 tabs open in 6 different windows!!! 

My eyes and head are aching from all the knowledge and experience on these 2 models within this forum!!!

I don't need 2 channels, and desire the lower THD. So, AGAIN, I digress... The SDG805 is the top contender!!! 

[ebastler]

...
I also WELCOME any suggestions or even questions concerning my use of the proper generator.
1) Which would suit my needs best? (Injecting an accurate low-distortion sine wave into audio gear)
...

I am still a bit unclear about your intended use of the generator. Asking for a very-low-THD signal to inject into an audio amplifier only seems to make sense if you also have a way to measure the THD at the output, but that side of the measurement chain did not get much attention here. Someone mentioned a Panasonic generator & analyser (which might well be outside your budget?) in an earlier post, but the discussion has gone back to focus on the generator side only.

For audio work on a budget, my bet would still be a decent USB sound interface with one of the various shareware programs that let you measure frequency sweeps, THD, FFT spectra etc. by using the sound card to both generate and measure your signals. I have used AudioTester (http://www.audiotester.de) for many years; on another thread someone mentioned Visual Analyser (http://www.sillanumsoft.org - I don't have personal experience with that one). For tube amp work, you need to be careful with the input voltages, of course; maybe add a homebrew switchable input attenuator.

Complement this with a cheap, battery-powered signal injector (with bad THD) for those instances where firing up the PC is too much hassle: E.g. following a signal path for simple good/no-good testing. -- Just my 2 cents. Best of luck with your audio repair and tuning work!

[oldway]

To test and adjust tube amplifiers, a sine wave generator with 0.3% THD is sufficient.

Most of the tube amp have a THD of 1% or more.

As a first approximation, you can subtract the sine generator THD from THD measured at the output of the amp to find the THD from the amp.

High precision THD measuring is nonsense with tube amp.

(NB: I use a Philips PM5109 RC Generator 10 to 100Khz and HP 331A distortion analyser.
Note that this last device also gives wrong readings because it is based on average values and not the RMS values of the signals.)

If you are looking for low THD, you better forget the tube amp.

But in reality, very low THD is not that important: a loudspeaker produces a distortion of the order of 2%, and even those who have a very good ear will only notice THD above 3 or even 5%.

[dom0]

Excellent speakers go below that level and headphones are generally much better. Another large issue with speakers is their strong intermodulation.

[oldway]

This information came from G.A. Briggs (Wharfedale) “Loudspeakers”.
Unhappely, i don't have this book no more so I can't say the page where he wrote this.

[dom0]

I'm not objecting you here ; even the best speakers have a linearity magnitudes worse than most solid state amplifiers.

[allen.gordon]

I would like to thank EACH of you for your input! It is greatly appreciated!!! 

I have downloaded and am currently using the VA software. It is an invaluable tool, and I couldn't be more happy with this discovery using your assistance!

I'm not trying to reinvent the wheel with tube amps, but I don't want to purchase a purpose-built generator when a multi-use generator is readily available.

At this time, I'm building single channel tube amps and gain-based effects pedals. I would like to expand my repertoire in the near future with the implementation of channel/tone/rectifier switching in tube amps, and time-based effects pedals.

I'm also working with my father on a portable sawmill he's built. It's is powered by hydraulics and we currently have solenoids for electronic control of the equipment. We have been self-discovering micro-controllers with decent success. The sawmill can now cycle through short programs and process an entire log (once the touch-off has been established).

Is there an inherent danger in purchasing overkill equipment that I'm not understanding? 

I absolutely LOVE that you are expressing a concern for spending money unnecessarily!  I appreciate that tremendously! I am definitely a fan of the "less is more" approach.

However, with the advent of nano-technology, micro-technology is at lower prices the EVER (of course out-sourced manufacturing also has a major affect on price points)! Since I'm in the position to make a modest investment in several pieces of test equipment, I'd like to future-proof my purchases by attaining the best technology for the best price.

Thank you again for all of your responses to my posts!

This is an awesome forum, filled with great and knowledgeable people! I spend MOST of my free time researching the forum and watching YouTube videos of Dave on the EEVblog channel!

Cheers,
Allen

[tautech]

Is there an inherent danger in purchasing overkill equipment that I'm not understanding? 

Not with the choice you have made.
Now you've shared a little more, you'll find an good quality AWG will be a much better choice for versatility.

[dom0]

I never really got the obsession with AWGs. 

They certainly have their uses, but I don't see any in the usual low profile audio-video testing and development. Well, okay, if you do analog video you could use an AWG to generate video, but then again any video generator does an easier and better and still cheaper (no one wants them anymore) job at that...