fmak said:I have been trying to get access quickly to the HP339A improvement article by Charles Hansen in the March 03 issue of AudioXpress. Unfortuntely one web listing in uk appeared wrong, the issue for purchase being Feb 03 but with the March 03 content.
Charles, or anyone else, can you pl help?
The HP product was solid to begin with and does not need any POOGE monkey business. All he does is replace the 80 KHz lowpass with a steeper <30khz board, assuming you won't need this for broadcast work and you will want to use it on CD source.
Re: Re: HP339A article Charles Hansen
Having read the article, I tend to agree. Don't see the point of a 20kHz brickwall filter on top of the CD brickwall filter.
Now a mod to extend dynamic range, reduce residual noise in the instrument will be really worthwhile; perhaps by replacing some of the opamps.
----------------------------------------------------------------------------slowhands said:
Having read the article, I tend to agree. Don't see the point of a 20kHz brickwall filter on top of the CD brickwall filter.
Now a mod to extend dynamic range, reduce residual noise in the instrument will be really worthwhile; perhaps by replacing some of the opamps.
Re: Re: Re: HP339A article Charles Hansen
It's kind of silly, but the real point is to make cheap $99 CD players have better specs than they deserve. Some cheap CD players have poor filters that let through a lot of noise, or they use noisy op-amps. Adding another brickwall filter in the measuring device willl "fix" the numbers for better sales.
If you don't believe me, take a look at the data sheet for nearly *any* audio DAC chip. They will have pins for "Left Mute" and "Right Mute". These will output flag signals when the data on the disc is all zeros. The Japanese manufacturers connect these flags to transistors switches that short the output of the analog stage. This gives incredible S/N ratio measurements on the test bench, even when using noisy op-amps and poor power supplies. Of course it has nothing whatsoever to do with real-world performance....
fmak said:
It's kind of silly, but the real point is to make cheap $99 CD players have better specs than they deserve. Some cheap CD players have poor filters that let through a lot of noise, or they use noisy op-amps. Adding another brickwall filter in the measuring device willl "fix" the numbers for better sales.
If you don't believe me, take a look at the data sheet for nearly *any* audio DAC chip. They will have pins for "Left Mute" and "Right Mute". These will output flag signals when the data on the disc is all zeros. The Japanese manufacturers connect these flags to transistors switches that short the output of the analog stage. This gives incredible S/N ratio measurements on the test bench, even when using noisy op-amps and poor power supplies. Of course it has nothing whatsoever to do with real-world performance....
the HP339A oscillator is perhaps the limiting factor -- it's distortion is about 0.0018% -- as if this were a limiting factor anyway, but there are plans for oscillators with distortion two factors of magnitude lower on the Linear Tech website --
as far as "utility" the nice big mirrored analog meter of the hp339 allows you to sweep a control setting on the DUT until you've got the result you want. A digital display (as in the Krohn Hite or Tektronix instruments ) requires some mental gymnastics to derive the same result.
as far as "utility" the nice big mirrored analog meter of the hp339 allows you to sweep a control setting on the DUT until you've got the result you want. A digital display (as in the Krohn Hite or Tektronix instruments ) requires some mental gymnastics to derive the same result.
jackinnj said:the HP339A oscillator is perhaps the limiting factor -- it's distortion is about 0.0018% -- as if this were a limiting factor anyway, but there are plans for oscillators with distortion two factors of magnitude lower on the Linear Tech website --
------------------------------------------------------------------------------
The limiting factor is actually the filter. There is residual of the fundamental around 93 dB down at 1kHz. The sig gen is around -96 dB down from 1 V.
I bought a machine in calibration for £300. Beautiful inside, very used outside. Compared to my National VP7xxx analyser, the HP is clearly inferior in performance and in ease of use. My problem is that the National is malfunctioning some of the time!
Anyone know where I can get knobs for the HP339A?
fmak said:You noticed that the knobs crack easily -- same on mine!jackinnj said:the HP339A oscillator is perhaps the limiting factor -- it's distortion is about 0.0018% -- as if this were a limiting factor anyway, but there are plans for oscillators with distortion two factors of magnitude lower on the Linear Tech website --
------------------------------------------------------------------------------
The limiting factor is actually the filter. There is residual of the fundamental around 93 dB down at 1kHz. The sig gen is around -96 dB down from 1 V.
I bought a machine in calibration for £300. Beautiful inside, very used outside. Compared to my National VP7xxx analyser, the HP is clearly inferior in performance and in ease of use. My problem is that the National is malfunctioning some of the time!
Anyone know where I can get knobs for the HP339A?
Dear all,
Sadly, I managed to connect the 350Volt +V of a tube amp to the output of my HP339 distaortion analyzer oscilator..
I am afraid my pride&joy , this beatiful machine is now dead.
Has anybody here ever tried to reapir such a unit.. ? Any hints or experients with repairing a HP339 te share?
grtz,
Thijs
Sadly, I managed to connect the 350Volt +V of a tube amp to the output of my HP339 distaortion analyzer oscilator..
I am afraid my pride&joy , this beatiful machine is now dead.Has anybody here ever tried to reapir such a unit.. ? Any hints or experients with repairing a HP339 te share?
grtz,
Thijs
It's kind of silly, but the real point is to make cheap $99 CD players have better specs than they deserve. Some cheap CD players have poor filters that let through a lot of noise, or they use noisy op-amps. Adding another brickwall filter in the measuring device willl "fix" the numbers for better sales.
If you don't believe me, take a look at the data sheet for nearly *any* audio DAC chip. They will have pins for "Left Mute" and "Right Mute". These will output flag signals when the data on the disc is all zeros. The Japanese manufacturers connect these flags to transistors switches that short the output of the analog stage. This gives incredible S/N ratio measurements on the test bench, even when using noisy op-amps and poor power supplies. Of course it has nothing whatsoever to do with real-world performance....
Hi Charles,
I agree. THD measurements with very limited bandwidth are of limited use. To some extent, single-number THD specs at limited bandwidth (or at 1 kHz) are part of the reason why some dismiss THD as being of little use. Doing 20 kHz THD on an amplifier with 200 kHz analyzer bandwidth can be quite revealing. Even better, if the residual is sent to a spectrum analyzer to enable viewing the individual harmonics. Alas, there are not many good audio spectrum analyzers out there that go above 100 kHz.
For these reasons, my favorite HF test these days is 19+20 kHz CCIF with spectral analysis (like John Atkinson does). The beauty of this is that the stuff you want to see lies in-band. BTW, this is especially important for measuring distortion on class D amplifiers, whose output filters will mask most of the interesting THD spectra.
BTW, 19Hz + 20 Hz "CCIF", with a 1 Hz beat note, can be a real LF torture test for a power amplifier.
Cheers,
Bob
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.