Topic: zoom h2

[Roving Sign]

got the replacement H2 today...1st one had only left channel on line-in.

well, i tested the line in now on the new H2 with my mics+preamp, all fine.

then i tested it with my bbox+mics =>bingo, same fault again.

seems my SP-SPSB-9 12V bbox is killing the right channels on edirol r-09 and zoom H2s......
after that i tried it 1 more time with my preamp, same fault...only left channel recording.

had no problem with this bbox on my former iriver and my sony mz-rh1 recorders.

IF my 12v bbox causes the fault, HOW the hell can that happen technically?

well, i won´t send it in for replacement a 2nd time...my intention is to use the internal mics only.....

Get a voltmeter and see if your SP battery box is OUTPUTTING DC into the right channel. 

Just connect the voltmeter to ground and the output pin of the box's output connector (not connected to anything else) 

It should measure at or near 0 volts.  Some decks have a DC decoupling capacitor that helps protect the input from applied DC, but maybe H2 has too little protection if given too much unexpected DC power input?

i just measured the output with a connected 3.5mm cable plugged in.
the TOP-ring of the plug shows: 150mV
the middle ring shows: 80mV

Well, the plot thickens as NO help from this. 

Middle ring usually the right channel, which is lower than tip-left channel voltage.  Either is low enough to only cause a pop sound in the audio, never a 'smoke' problem to anything.

SO MAYBE THIS IS A MECHANICAL STRESS ISSUE like R-09?  Usually by wiggling a broken input you can sometimes get the audio back if holding plug in some position.  Maybe try this with phones on?

If you can hear the audio come back momentarily in the phones while in REC/standby, then mechanical failure (PCB board/input connector) is most likely.

Longshot - arent there some 1/8 connectors that are a little bigger than most?

I know I've seen some Switchcraft stuff that **LOOKS** like an 1/8 mini - but is just a tad bigger...

Hard to believe 1 user could have the exact same problem across two completely different brands...

[Arni99]

it was the 1st time i used the new H2 today.
1. tried it with my preamp which showed both channels recording.
2. then as described above, i unplugged the preamp and connected my bbox with the right angle mini-plug to the line-in=>1 channel gone forever.

i can exclude the stress-issue, my RH1 is the best piece i ever had (2 r09, 2 iriver H120, 2 zoom).
the irivers and the RH1 never had a problem with my bbox and my 3.5mm right angle plugs are from SP.
anyway...I´m soon off to record Brian Auger´s Oblivion express tonight with my RH1+DPA 4061+my killer battery box .

[guysonic]

WHILE HARD TO TELL FOR SURE, the plug with two cords entering looks suspect to be larger less precision type with shearing ability 'ledge' type ring.
These can as mentioned cause some jack types real grief, but consistently with just one insertion?  That is hard to believe, but is possible I guess.

[Arni99]

WHILE HARD TO TELL FOR SURE, the plug with two cords entering looks suspect to be larger less precision type with shearing ability 'ledge' type ring.
These can as mentioned cause some jack types real grief, but consistently with just one insertion?  That is hard to believe, but is possible I guess.

the 2 cord plug was from my sp-cmc-4U mics and i never plugged this into the edirol r09 or the zoom.
just wanted to show i use standard SP 3,5mm plugs .
i guess it MUST be my 12v bbox output which causes some elctronic fault to r09 or zoom H2 recorders.
my h2 will return tomorrow and i hope i´ll get my money back, the shopp offer´s "30 day money back"....

[guysonic]

Found time to at least start the testing of H2.  As found with H4 model, H2 has #100-#127 working rec level control range.  H2 seems not to have weird rolled off frequency found in the H4 model I tested previously; www.sonicstudios.com/zoomh4rv.htm

While the internal mics seem low enough noise to be useful even for natural acoustic sounds, connecting up external mics seems plagued with excessive low frequency and high switching spike noise as the input noise plots I ran indicate.  Mic input -3 dB bandwidth is 50-40,000 Hz.  So low frequency is being rolled off maybe too much for some acoustic recording purposes. 




Line input -3 dB response is 10 - 42,000 Hz and is most useful with external preamplifier with attenuation network to reduce signal to <-1.9 dBu H2 maximum allowed.





These plots and more H2 technical info to be posted sometime on my site in an upcoming review of this lowest costing feature-loaded deck.

[illconditioned]

Thanks Guy.

I appreciate all your testing.  As I've said in other threads, it pays to be paranoid about this stuff.  Just like computing, both hardware and software, there are so many products out there that just do not work.

  Richard

[Church-Audio]

Found time to at least start the testing of H2.  As found with H4 model, H2 has #100-#127 working rec level control range.  H2 seems not to have weird rolled off frequency found in the H4 model I tested previously; www.sonicstudios.com/zoomh4rv.htm

While the internal mics seem low enough noise to be useful even for natural acoustic sounds, connecting up external mics seems plagued with excessive low frequency and high switching spike noise as the input noise plots I ran indicate.  Mic input -3 dB bandwidth is 50-40,000 Hz.  So low frequency is being rolled off maybe too much for some acoustic recording purposes. 




Line input -3 dB response is 10 - 42,000 Hz and is most useful with external preamplifier with attenuation network to reduce signal to <-1.9 dBu H2 maximum allowed.





These plots and more H2 technical info to be posted sometime on my site in an upcoming review of this lowest costing feature-loaded deck.

Just so you understand you cant really do noise tests like that.. You must use averaging in order for your noise measurements to make any sense.. Here is a real noise floor test. You cant just take a peek level and then say well there is a peek at -98 so thats the noise floor it must be averaged. What I do is take a 1k signal apply it to the device undertest and then use a 1k filter and subtract the 1k signal what your left with is the noise floor as can be seen on the grey screen that says "in band level" Notice my noise measurements for the zoom H4 in the graph below it shows an inband level of -103db that is the noise floor. I also think your elevated low end frequency response is a sample rate error for the device under test. With the zoon you must adjust your sample rate of your program and sync it with the sample rate of the zoom in order to get a real frequency response because the zoom has a major sample rate error its off big time from my computer. When I first did the tests I was getting similar results until I adjusted my sample rate of my computer by .05% then they synced up. BTW my noise test was A weighted so Its actually more like a noise floor of -97 or so.


Chris

[guysonic]

More H2 info showing effects of having the backlight on/off.  NOT much difference except just a little more noise around 800 Hz if using MIC input.  LINE input seems unaffected either way.

Like all my spectrum tests, the FFT is set at maximum number of samples with the settings showing in the image below, but same as all noise tests done, and all using 1000 ohm metal film .5% input load with deck recording inside a metal box to exclude external influences.  Recorded files on the flash card are later transferred to CEP for FFT noise analysis plots like you see posted.




Someone living in ZOOM land argues for 'A' weighting on 'frequency plot' (spectrum) noise measurements.  Maybe done differently in Canada, but 40 years experience in USA and never seen 'A' weighting used on noise spectrum graphed over frequency, 'A' mostly for arriving at a single dB number assumed to be humanly audible under certain conditions, and has meaning for physiological effects. 

'A' weighting is based on the Fletcher Munson curve that rolls off low and high frequency bands to approximate audibility.  As most will realize when viewing the F-M graphic, the low-to-high wiggle curve gets way more straight as the loudness increases. 

So 'A' weighting is not meant to be precise, just an approximation finding best use in Sound Level Meters giving some measure of acoustic loudness for possible health effects, but little usefulness in frequency spectrum plots where nothing should be hidden giving chance of unassumed performance assessments. 

FAST EXAMPLE: An 'A' weighted instrument will not register highly audible quick impulse sounds like gun shots (especially at some distance), and spark gap sounds are invisible to 'A' weight readings, but VERY visible to what we hear.  So 'A' is good for getting at long term physical effects, but NOT for knowing what we hear.   Marketing an 'A' weighted S/N does give nice low number to display.  It may be 'non'-weighted S/N specifications number tells a less biased story?     

What would you prefer to see, -92 dB S/N A weighted, or -85 dB unweighted spec for the same product?  And how would you compare this products -92 dB "A" spec with -90 dB  unweighted device.  Easy with knowing -85 dB non-'A' noise figure to see -90 dB device has less mostly audible noise , but only seeing typical 'A' weighted spec on one device, no easy apples-to-apples comparison. 

Even with both products displaying weighted OR UN-weighted numbers, only a psychic could know the audible noise 'characteristics' of these two devices. 

It would seem best chance of actually seeing real differences of noise characteristic is by unadulterated signal level verses frequency graphics like I've been using in the reviews. 

With spectrum graphics, just by looking at dB levels verses frequency to get a real feel for noise performance.  And having spectrum graphics for many new models all tested in an identical manner allows uncommon 'apples-to-apples' comparisons. 

What's not to like with this?

[Church-Audio]

Someone living in ZOOM land argues for 'A' weighting on 'frequency plot' (spectrum) noise measurements.  Maybe done differently in Canada, but 40 years experience in USA and never seen 'A' weighting used on noise spectrum graphed over frequency, 'A' mostly for arriving at a single dB number assumed to be humanly audible under certain conditions, and has meaning for physiological effects. 

With spectrum graphics, just by looking at dB levels verses frequency to get a real feel for noise performance.  And having spectrum graphics for many new models all tested in an identical manner allows uncommon 'apples-to-apples' comparisons. 

What's not to like with this?

Why do you have to Bring CANADA into this? it really shows a lack of class on your part. And thats too bad.


I was not saying that A weighted measurements were the "way to go" I was saying that alot of companies are using them. Where your test is really flawed is the fact that you dont average out your noise measurements. That's the only problem I had with your measurement. You cant "eyeball" a graph and say yep that's a noise floor of -150 buy just looking at it. Sooner or later you have to put it into a single number. And terminating your input does not guarantee that noise from your environment is not getting into the DUT.

BTW Neumann, Sennhieser, Rane, Buzzaudio, M-AUDIO..And there are many more...They  ALL USE A WHEIGHTED MEASUREMENTS... but hell what do I know.. Now again I am not saying they are correct in doing so. But your theory of nobody using A-weighted measurements seems WRONG. Because they are.. And they will continue to do so whenever a product does not meet the magic number weighted measurements will always be used to trick the consumer into seeing better performance. Now when you dont Average a noise spectrum ( the main mistake your making with Your measurements ) you are making huge mistake.. Please tell me if you can what the graph below indicates for a noise floor..

[guysonic]

Terminating the input and placing the device under test in a solid metal box (like I stated doing for all such tests) DOES eliminate chance of outside interference.   

Obscuring details of noise floor over frequency with giving a single unqualified graph (no test conditions stated) or single boiled-down number only aids in covering up real performance details so NO WAY anyone can tell what is going on with each device (like you're insisting) is really wrong way to do honest testing from my experience, but great way to cover up seeing good performance and shortcomings.

Curious that you only pop-up with this stuff when I review a ZOOM product, but seem to not notice or present such arguments with accompanying obscure graphs when posting MT1, MR-1, or R-09 tests and resulting reviews. 

I am now thinking maybe you're living too close to ZOOM company for being objective about revealing type testing of their products?

[Church-Audio]

Terminating the input and placing the device under test in a solid metal box (like I stated doing for all such tests) DOES eliminate chance of outside interference.   

Obscuring details of noise floor over frequency with giving a single unqualified graph (no test conditions stated) or single boiled-down number only aids in covering up real performance details so NO WAY anyone can tell what is going on with each device (like you're insisting) is really wrong way to do honest testing from my experience, but great way to cover up seeing good performance and shortcomings.

Curious that you only pop-up with this stuff when I review a ZOOM product, but seem to not notice or present such arguments with accompanying obscure graphs when posting MT1, MR-1, or R-09 tests and resulting reviews. 

I am now thinking maybe you're living too close to ZOOM company for being objective about revealing type testing of their products?

LOL Yeah I work for ZOOM, man you got me. Hey I should warn you that little green men are at this very moment waiting to break into your test bench and install some real software on your computer.
Lets just stop nobody wants to see a flame war. And besides I have alot of orders to fill. 

 

[guysonic]

Chris's graphics posted in this thread, and only partially explained testing criteria is more visible in his previous post about H4 model found here: http://taperssection.com/index.php/topic,82333.msg1092560.html

[Church-Audio]

Chris's graphics posted in this thread, and only partially explained testing criteria is more visible in his previous post about H4 model found here: http://taperssection.com/index.php/topic,82333.msg1092560.html

No offence but I did not explain my testing procedures only because I thought who cares? your tests are flawed. I proved that. Now do I like the Zoom No I think its a toy. Do I work for Zoom no. Do I care what you think? No.. so why not just drop this. I got different results then you because I am not using a sound editing program for sound analysis I actually whent out and spent money on a real software package that analyzes audio for me. Every test I do is first done by doing a loop back test from my test interface's input to output. Then I take an output from my interface and then I record on the DUT. Then I take the file from the DUT and analyze it.. In my program.. Then I take the device and do a loop back test thru the device it self to see how device performs under these tests.. I dont use a Faraday shield because these devices should not be tested in one.. Unless the people using the devices are going to be using a Faraday shield when they operate it. Your tests dont prove the real performance of the DUT. That's what I was saying now I am done with this convo. Please lets move on.
One last thing you still have not answered my question if your methods of eyeballing noise floor are so good please tell me what the noise floor is in this graph..
Chris

[guysonic]

   Maybe the difference is I look at this more from a Tapers viewpoint; this is how I began many years ago, and am still following my passion for recording all kinds of sounds.

Yes it's true tapers don't use their decks inside a metal box, and neither are they doing I/O signal loops!  

  I test just like tapers use them.

But for me, the Faraday shield box stays with the program as external noise has no place being added to recordings of deck's self-noise.  External noise susceptibility is a different test done ONLY after knowing self-noise reference levels.

My testing ONLY involves analog input-to-recorded file system performance.  This means working the test like the deck is used and analyse the resulting recording in a simple audio editor.  Nothing wrong with this, and many tapers can do such tests these days themselves.

AND PLEASE, NO technical mumbo-jumbo showing huge (good only for marketing) numbers from inputs unknown, and invisible signal conditions, and defer weighted numbers for marketing purposes.  

Tapers and technical people like myself get no benefit from pseudo-technical obscurity working against understanding what's going on.  We need to know in input/settings practical terms and definitive spectrum graphics ways to get the best experience from using a particular model of portable deck.  

   So doing real input testing while recording and then showing definitive graphics as I've done seems most useful from a tapers viewpoint, and (most) technical people also appreciate this testing technique for being a most concise way to understanding audio deck performance.

I am not saying Chris is doing anything wrong, like he's claiming for me, and he can use whatever methods, graphic displays, and explanation, or lack of as he desires in my book.   I personally find his technical displays impossible to understand in any real taper related terms.  

But maybe others better appreciate, getting benefit from his tech posts.   I encourage Chris to keep at it as there's always chance he will improve his writing ability with practice so even I can get some benefit.  

However, I do object to his statements of what I'm doing is invalid !!  

This seems unnecessarily asking for trouble where there's no technical, moral, or ethical problem with my reviewing methods that I'm aware.  And I've been doing this type of testing for over 40 years without complaints.


 

[Church-Audio]

But maybe others better appreciate, getting benefit from his tech posts.   I encourage Chris to keep at it as there's always chance he will improve his writing ability with practice so even I can get some benefit.  

POST EDIT.

I have decided to edit my post and remove it. I am doing this because I just dont want to clog a thread with this crap. I apologize to the people that were trying to read this thread and actually learn something from it.


Chris