Taperssection.com

Gear / Technical Help => Recording Gear => Topic started by: DSatz on November 25, 2018, 10:42:56 PM

Title: little Tascam single-channel recorder (DR-10X)
Post by: DSatz on November 25, 2018, 10:42:56 PM
Hi. Just in case anyone's interested: I will have a live recording situation later this week for which I'll need to place two spot mikes in hard-to-reach locations. Running cables from those mikes to my recording position, or even to a separate two-channel recorder of their own, would be unsightly, dangerous and time-consuming--so I plan to attach a miniature single-channel recorder to each microphone, and sync everything up afterward to the tracks from the main recorder. If the tracks from the spot mikes need time stretching or squeezing to sync up and stay "sunc", so be it.

The two recorders that I looked at were the Tascam DR-10X and something called the Saramonic SR-VRM1 (which I'd never heard of before). Both are in the $100-$125 range. Each has a balanced, XLR-3F mike-level input, and records just a single channel of audio.

The "Saramonic" recorder has built-in phantom powering, so I considered it first. I went to a store (Adorama, not far from Union Square in Manhattan) that had one that I could look at. The salesperson graciously allowed me to install batteries, power it up, and attach a Schoeps PHS 48 phantom power tester that I'd brought along. Sorry to say, the indicator LED on the tester didn't light up. I can only assume that the current draw was too much for the phantom supply circuit in the recorder. So, while that recorder had several advantages (a metal housing, finer control over level setting, and quite possibly a quieter preamp), I let it go and bought a Tascam DR-10X instead; I'll pick up the second one tomorrow.

The Tascam DR-10X is surprisingly small--only slightly bigger than a small bottle of hand sanitizer. But it was clearly designed for dynamic rather than (professional) condenser microphones. It lacks phantom powering, and an input of just 2.5 mV drives it to 0 dBFS at the lowest of its three available microphone gain settings. Professional condenser microphones often put out 40 mV or more for loud sounds, so I will have to "pad down" the input by at least 10 dB.

The recorder doesn't have any finer adjustment for its record levels than three discrete "LO/MID/HI" mike gain settings that are some 6 to 10 dB apart from one another--and those settings are made via menu items that can't be accessed while the recorder is recording. So that isn't a very wonderful arrangement. But its coarseness is compensated in part by the recorder's ability to record a second, "safety" track either 6 dB or 12 dB below the level of the main track. Again, that isn't useful if overload is occurring at the input of the preamp stage of the recorder; then you would only be recording your distorted signal at a lower level. But if there's enough resistive padding between the phantom power supply and the recorder's input, it should work.

(please note correction at end of paragraph) The recorder has 24-bit recording as a menu option--in fact, it's the default setting. But even at the lowest gain setting, the built-in mike preamp has a noise floor about 72 dB below full scale, i.e. only about 12 bits of actual resolution. (Of course that is as good as the best 15 or 30 ips analog open-reel recorder ever was--but nowadays we are spoiled.) So I set it for 16-bit rather than 24-bit recording. I made test recordings both ways, and the setting made no observable difference at all to the noise floor of the recording--even on the secondary track that was 12 dB lower. Correction added next day: A better way of reporting the same result would be to call it a nearly-80 dB dynamic range (based on RMS levels of signal and noise); the conclusion re: 16 vs. 24 bits is still the same, though.

The recorder runs on a single alkaline AAA cell for up to 10 hours. It has a low-cut filter at 120 Hz (slope not stated, and I didn't measure it) and optional settings for automatic level control and/or a limiter (neither of which I tried).

After I've had a night's sleep, I'll see what the dynamic range situation is when the recorder is set for what amounts to a line-in situation, which I could use by substituting Sound Devices MP-1 preamps for the battery-run phantom power supplies that I was going to use. According to the specs in the manual, that might give a quieter result; I doubt that it will be drastically different, but I'll see.

--best regards
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: jerryfreak on November 26, 2018, 12:44:00 AM
how about the single channel lectrosonics pdr+schoeps cmr
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: Sebastian on November 26, 2018, 04:24:04 AM
I'm a little confused. You didn't get the Saramonic that "had several advantages" because its phantom power supply wasn't up to spec. Instead, you got the Tascam that does not even have phantom power. May I ask why you made that decision?
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: Ozpeter on November 26, 2018, 04:52:38 AM
Come to think of it, that's a good question!  As for the phantom power - you did turn it on in the menu?  (Item 7) - ok, ok, just asking...
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: morst on November 26, 2018, 02:17:30 PM
how about the single channel lectrosonics pdr+schoeps cmr
Wireless might be even more hassle, and is probably more costly than a couple of cheap Trashcan recorders.


But it WOULD save the squash/stretch part of editing, as long as signal reaches its destination.


I guess it depends on how radio frequencies carry in the venue, and how many wireless signals are already present.
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: DSatz on November 26, 2018, 06:07:34 PM
Re: wireless: I've worked with a wireless mike for some documentary recording that I used to do every summer in recent years, and I had a pretty nice one-channel Sennheiser setup. Unfortunately, when the FCC decided to auction away more of the spectrum a couple of years ago, they auctioned away the range of frequencies that my setup operated in, and it can't be retuned. To replace it would cost a little more than $600 per channel and I wouldn't have any other foreseeable uses for the setup.

Plus the concert will be in midtown Manhattan where other wireless mikes are in heavy use all the time, including specially licensed users with higher transmitting power than general users are allowed to use. If another nearby transmitter were to start using either of my chosen channels during the concert, I couldn't do anything about the interference.

Finally, the inherent dynamic range of a wireless channel is quite limited, so all the systems use wideband companders. Those are optimized for speaking voice, and they do very well at handling that without audible artifacts; I've never seriously tried recording wide-range classical music through one, though, and I really don't know whether it would sound good or not. So I didn't want to go in that direction.

--The Lectrosonics PDR would certainly have been a high-quality alternative choice. And the option of using it with a Schoeps CMR is definitely appealing. I frankly had forgotten about it, though, and I'm not sure I could have pulled together and tested the complete setup for two channels with as little lead time as I have for this recording. Plus, $1000+ saved is $1000+ earned, to update an old saying.

--OK, why I didn't buy the Saramonic: because I had never heard of the company before--and especially once I saw that they didn't implement phantom powering properly, I didn't trust them enough. Just in terms of playing the odds for reliability, I would rather buy from a well-known company that I've dealt with for decades. Tascam gear in my experience may not always have the very best audio quality, but it generally holds up well in rough use, and can be fixed if it breaks. (And yes, I did enable phantom powering in the recorder's menu, and the red LED came on. Don't feel bad about asking--"senior moments" are definitely a factor in my daily life nowadays.)

--Finally, I tested the "EXT" setting for the input, which is quasi-"aux" level. It gives a 10 dB quieter recording, so that's what I'll use. I still don't see a use for the 24-bit setting on the recorder, though; next chance I get, I'll look more carefully at the dither situation for the safety track (in dual recording mode) to see whether it improves low-level linearity or not.
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: jerryfreak on November 26, 2018, 06:15:54 PM
how about the single channel lectrosonics pdr+schoeps cmr
Wireless might be even more hassle, and is probably more costly than a couple of cheap Trashcan recorders.


lectrosonics pdr is not wireless its a professional flash recorder

https://www.trewaudio.com/product/lectrosonics-pdr/

its the single channel version of the SPDR 2-channel that has been discussed recently
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: morst on November 26, 2018, 09:33:50 PM
Unfortunately, when the FCC decided to auction away more of the spectrum a couple of years ago, they auctioned away the part that my setup operated in, and it can't be retuned. To replace it would cost a little more than $600 per channel and I wouldn't have any other foreseeable uses for the setup.Plus the concert will be in midtown Manhattan where other wireless mikes are in heavy use all the time, including specially licensed users with higher transmitting power than general users are allowed to use. If another nearby transmitter were to start using either of my chosen channels during the concert, I couldn't do anything about the interference.

lectrosonics pdr is not wireless its a professional flash recorder

https://www.trewaudio.com/product/lectrosonics-pdr/ (https://www.trewaudio.com/product/lectrosonics-pdr/)

its the single channel version of the SPDR 2-channel that has been discussed recently



Ah thanks for the clarification jerryfreak.


And DSatz, I know what you mean about the FCC squeezing the space, and the unreliability of trying to work in a wireless jungle.


I wonder if you could sell that gear that no longer can be run in USA via ebay or reverb.com to someone in a country which does not observe the same FCC restrictions?


I'm curious if guitarists or wind players who use wireless setups do anything special to get better quality than vocal wireless systems offer. I guess they are mostly compromising audio quality for convenience/practicality/ability to do things you could not do wired? If they want quality they can go into the studio...
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: 2manyrocks on November 29, 2018, 05:36:25 PM
The DR-10x is actually a popular option for video shooters who don't trust wireless and/or don't want to spend the $ on wireless.  BH puts a white version on sale every so often for $129 if I remember right. 
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: DSatz on December 02, 2018, 04:49:18 PM
In the end, I made the recording without the spot mikes. But before putting these two little recorders away, I decided to make a quick and dirty test by using a Y-cable to feed them both from the same test signal generator. I recorded about 76-1/2 minutes of test signal at 48 kHz. The two recorders' clocks drifted apart by about 10.5 samples per minute on average, or about 7 milliseconds across the whole interval. If I'm counting on my fingers correctly, that's about a 3.5 parts per million difference.

Adobe Audition CS 6 is now working to adjust for this difference as well as it can; I have a rather fast Intel i7 computer, and the wait cursor says that it will take more than half an hour. The "Stretch and Pitch" effect has a precision of about 1 millisecond for its settings, so when it is done, the two tracks will still end ~100 samples apart. That's better than ending 760+ samples apart as they were originally, but I don't have any way to align them more precisely.

So I'm glad I didn't use this approach. It probably could be made to work with a lot of time and effort, but I don't have time like that. Or maybe there are better software tools available for this specific task, I dunno.

Of course I'm also assuming (tacitly until now) that both clocks ran at highly constant speeds that were just a little different from one another. If they were also unsteady, that can't be fixed afterward. (Interestingly, that problem actually can be fixed nowadays in analog recordings. There is usually some remnant of high-frequency bias signal on the tape, which can be extracted and used as the basis for reclocking the recording computationally. For those interested, see https://www.plangentprocesses.com/ .)

TL;DR is therefore, "Kids, don't try this at home."

--best regards

P.S.: 2manyrocks, Tascam makes (and B & H sells) an alternate version of this recorder called the DR-10L. It's wired for use with lavalier mikes via an unbalanced mini-jack input with "plug-in powering" and it is available in either dark gray or white for $200 including a lavalier mike. (The XLR model, without no mike included, costs only about $110.)
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: morst on December 02, 2018, 05:43:50 PM


I recorded about 76-1/2 minutes of test signal at 48 kHz. The two recorders' clocks drifted apart by about 10.5 samples per minute on average, or about 7 milliseconds across the whole interval. If I'm counting on my fingers correctly, that's about a 3.5 parts per million difference.


Adobe Audition CS 6 is now working to adjust for this difference as well as it can; I have a rather fast Intel i7 computer, and the wait cursor says that it will take more than half an hour. The "Stretch and Pitch" effect has a precision of about 1 millisecond for its settings, so when it is done, the two tracks will still end ~100 samples apart. That's better than ending 760+ samples apart as they were originally, but I don't have any way to align them more precisely.


So I'm glad I didn't use this approach. It probably could be made to work with a lot of time and effort, but I don't have time like that. Or maybe there are better software tools available for this specific task, I dunno.


Of course I'm also assuming (tacitly until now) that both clocks ran at highly constant speeds that just happened to be a little different from one another. If, instead, they are both unsteady AND on a long-term average, also slightly different from one another, that would be far more complicated to deal with.


TL;DR is therefore, "Kids, don't try this at home."






I do this at home all the time. Anytime the board feed arrives via a USB thumb drive, I get to indulge my spreadsheet. Check it out, the bottom line (20-21) is this project. Everyone should feel free to download this and customize it with their own projects (and delete mine or leave them as a guide)
https://docs.google.com/spreadsheets/d/1pQGfYwPgBFFzcY5m6aRj-Zbu9HsRumLy-tJB1d8Eufg/edit?usp=sharing (https://docs.google.com/spreadsheets/d/1pQGfYwPgBFFzcY5m6aRj-Zbu9HsRumLy-tJB1d8Eufg/edit?usp=sharing)




According to the math, if your 76.5 minute file at 48k is 220320000 samples, and you need to stretch the short one by 760 samples, I get a percent difference of  -0.00034495% (rounded a bit, since that's a lot of precision already, for a file of less than two hours)


I would use the free program Audacity's Speed Change plugin to sync 'em up, and it would not take 5 minutes on my 2011 macbook pro laptop with intel i5 at 2.3 gHz!


Free download for windows: https://www.audacityteam.org/download/windows/ (https://www.audacityteam.org/download/windows/)




As for the measurable/ philosophical / practical question of how the clocks vary, I usually select my sync points as far apart as I can get them for max precision. Remember if one mic moves during a recording, it throws the distance off, so you no longer have a fixed reference at all. I measured my pair of Sony PCM-M10's several years ago. I made a test recording of a ten second tone, followed by 1 hour of silence, then another 10 second tone. I loaded it on the memory card of one machine, then I played it from that one and recorded via analog input on the other. I did it three times and the files were different lengths. This leads me to think that it's dependent on temperature or something.
 
Jitter vs Drift... Jitter should be really small. Drift is comparatively large between any two clocks. I've found that I don't really notice a difference of up to about 15 ms but over that it starts to get weird. Your 760 samples out of 48000 is 0.0158333 seconds, or 15.83 ms. If this was rock & roll, you could get away with lining them up in the center, and letting it drift into then out of sync!
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: DSatz on December 02, 2018, 06:35:50 PM
morst, many thanks--I downloaded and installed Audacity. Its "Change Speed" plugin did the work in 16 seconds, and got the file to within a millisecond of the desired length. The deviation between tracks is now down to 23 samples after 76+ minutes.

As you say, one could sync them up in the middle of the file, and then the worst-case drift would be half of that.

Adobe Audition (which cost several hundred dollars) took over half an hour, got the requested length wrong by several milliseconds, and did some extremely weird stuff to the audio of the file--in what had been a long silent section (just the analog noise floor of the recorder), it created audio that sounds like birds chirping softly in a reverberant forest, then building up to several minutes of full-scale screechy noise of a kind that would make a dentist reach for hearing protection. This is with the "izotope" plug-in that Adobe considers to be their higher-performance plug-in, with "precision" set to the maximum value.

So, hmmmm. And again, many thanks.
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: morst on December 02, 2018, 07:00:18 PM
So, hmmmm. And again, many thanks.
Glad to help!!!
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: 2manyrocks on December 05, 2018, 10:22:12 AM
Sorry about confusing the model you were discussing.  I've been curious about trying the DR-10x with an EV 635 as an interview mic.  What do you think of that combo?

Interesting that Audacity keeps proving its usefulness and ease of use over more expensive options.
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: morst on December 05, 2018, 03:09:16 PM
Sorry about confusing the model you were discussing.  I've been curious about trying the DR-10x with an EV 635 as an interview mic.  What do you think of that combo?

Interesting that Audacity keeps proving its usefulness and ease of use over more expensive options.
635 or 635a? They are VERY different looking.
635 is a 1950's style mic that looks like a coffee can attached to a stand mount via a hinge.
on ebay one just sold for $50 that looked pretty clean, probably a good harmonica mic.



On a quick web search, I read that the EV635a is a "popular interview mic" in the news gathering field.
But it's an old model too. Early 1970's, I think.
It's omnidirectional, so that makes it easy not to have to anticipate who will be speaking so you have to point a directional mic at them.
It's a dynamic mic so it doesn't require phantom power.
Should be fine for talking, especially if you already have it. Also marketed as a PL5 in music stores, in a slightly different case. If you are buying one, ebay has listings up to $200 but the completed items show sales down to $30 for working ones.


Audacity is free. EVERYTHING is more expensive!!
 :yahoo:
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: DSatz on December 06, 2018, 01:30:08 PM
one last (I hope) footnote about the DR-10X: One of the biggest complaints in the reviews I read on line was that the metering turned itself off when the unit started recording. But that is no longer true with the latest firmware. You still can't access the menu while recording, and therefore can't adjust the input gain of the recorder itself. But now you can see what levels it's getting, which could be very useful, especially if you're driving the recorder from an outboard preamp that has adjustable gain.

By the way, the meters allow a small amount of unadvertised headroom. The last little indicator line lights up at ~2 dB below full scale.

--best regards
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: 2manyrocks on December 07, 2018, 07:13:13 AM
I have the 635A, the little interview mic.  Used to see them used on TV field recordings all the time.  Now reporters use their cell phones. 
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: morst on December 07, 2018, 03:11:43 PM
I have the 635A, the little interview mic.  Used to see them used on TV field recordings all the time.  Now reporters use their cell phones.
Test it out. The cellphone might be less noisy, just due to modern circuit design!?
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: IronFilm on December 11, 2018, 03:19:52 AM
The DR-10x is actually a popular option for video shooters who don't trust wireless and/or don't want to spend the $ on wireless.  BH puts a white version on sale every so often for $129 if I remember right.

That is the Tascam DR10L, which is a little different.
It is a bodypack recorder which comes with a lav.
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: jerryfreak on January 21, 2019, 07:01:40 AM
love the size of the DR-10L, too bad its not 2 channel
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: DSatz on February 02, 2019, 04:47:33 PM
Going back to the issue of clock drift between two outboard, mono recorders: The Lectrosonics recorders are specified to have better than 1 ppm clock accuracy by means of a temperature-controlled crystal oscillator ("TCXO") circuit. Tascam doesn't give any such specification for the DR-10 recorders, but the drift I observed between my two was nearly twice the maximum allowable difference between any two Lectrosonics recorders.

That's not apples vs. apples, though. In terms of clock rate variation, I don't know how typical my two Tascam recorders are of their species. Plus if these errors are randomly distributed, the discrepancy between the clock rates of two randomly selected Lectrosonics recorders should, in the great majority of cases, be distinctly less than the maximum allowable difference.

It's interesting to realize the degree to which even now, some 40 years into the digital recording "revolution", the clocks in professional recorders still can't stay in sync with one another even for modest periods of time. 1 ppm accuracy, if I'm figuring this correctly, means a potential deviation of about three samples per minute at 48 kHz--so two separate recorders each having 1 ppm accuracy could slip apart at the rate of one sample every ten seconds or so.

Of course, thinking back, the very best analog studio recorders (with crystal-controlled servo capstan motors and advanced tape tension controls on both the supply and take-up reels) specified 0.1% speed accuracy, a/k/a 1,000 ppm ... so maybe I shouldn't complain too loudly.

--best regards
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: jerryfreak on February 02, 2019, 05:49:01 PM
Going back to the issue of clock drift between two outboard, mono recorders: The Lectrosonics recorders are specified to have better than 1 ppm clock accuracy by means of a temperature-controlled crystal oscillator ("TCXO") circuit. Tascam doesn't give any such specification for the DR-10 recorders, but the drift I observed between my two was nearly twice the maximum allowable difference between any two Lectrosonics recorders.

at which point (in samples or ms) does clock drift become an audible issue? like if you drifted at maximum rate of either the lectro or the tascam, would there be a difference after an hour?

in other words, in absence of timecode, it would seem that you would either always correct/compress one stream to match the other, or almost never do it? i would think two identical modern devices with similar components would track pretty well from unit to unit
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: morst on February 02, 2019, 06:31:49 PM
at which point (in samples or ms) does clock drift become an audible issue? like if you drifted at maximum rate of either the lectro or the tascam, would there be a difference after an hour?

in other words, in absence of timecode, it would seem that you would either always correct/compress one stream to match the other, or almost never do it? i would think two identical modern devices with similar components would track pretty well from unit to unit
I generally allow +/- 20 ms max error, which would be +/- 960 samples at 48kHz. Some purists will say that they can hear 10ms smear but I don't find it much of a problem for rock & roll.


Sony, Tascam, Roland (Edirol)... none of these machines have tightly matched clocks between machines. After an hour, you can hear a "Flam" on almost any two recorders.
Use my spreadsheet if you would like to fiddle with some examples. You can download it if you prefer.
https://docs.google.com/spreadsheets/d/1pQGfYwPgBFFzcY5m6aRj-Zbu9HsRumLy-tJB1d8Eufg/edit#gid=583050244
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: jerryfreak on February 02, 2019, 09:43:54 PM
so these clock errors are just a bias and are replicatable?

like if clock a is 2ppm faster than clock b on a given recording, is that pretty consistent from recording to recording

temp appears to have some effect
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: morst on February 03, 2019, 03:10:38 PM
so these clock errors are just a bias and are replicatable?

like if clock a is 2ppm faster than clock b on a given recording, is that pretty consistent from recording to recording

temp appears to have some effect


They are not exact. I can't tell you that my system is perfect, but it gets me close enough. I think temperature does have some effect. When I first developed my spreadsheet and technique for sync, I ran three tests in a row with the same two recorders, and didn't get the exact same results each time. But it gave me a range.


(My initial test was done by creating a file made of a ten-second tone, 1 hour silence, then another ten second tone. I used the one hour silence as the timed test, and delineated it with the tones. I loaded it onto the card of one recorder, then played it back while recording via analog cable to the other machine. Three times.)
Title: Re: little Tascam single-channel recorder (DR-10X)
Post by: jerryfreak on March 14, 2019, 09:57:31 PM
I just snagged a DR-10C

(https://i.imgur.com/nr3prOp.jpg)

this is the version of the DR-10L that Tascam got stopped from selling in US due to Zaxcom's patent suit, apparently because it is designed to daisy chain in between mic and transmitter and record transmitter audio which zaxcom has a US patent on. (I'm guessing the DR-10L and the lectrosonics PDR/SPDR do not have line outputs to go to the next device and thats why they can sell them).

anyway the base device is the DR-10CS which comes with minijack in and out (TRS) which is setup for 2-wire sennheiser lav mics like the MKE. on the right is the adapter kit which replaces the top panel and hot plugs into the board to give you a 4-pin configuration compatible with Shure. the Shure adapter effectively makes it a "DR-10CH" which im not sure is an individual model they sell anymore. It looks like the "L" model avaialbel in the states has the same top cover though im not sure if the adapter plate would fit it (is there a modlar jack inside or is the L hard wired). also looking a tthe manuals the L and S have differnet firmware so im not sure if you could get an "L" to throw out 5V on the extra pin even if the board did attach

anyway this thing is TINY and can throw 2V of bias power and maybe run a dpa 4063. with the 4-pin Shure adapter plate there is a separate pin for +5VDC power which might be able to run a schoeps CMR, or potentially be jumpered over to the input circuit to feed a full 5V to a 4061.

https://tascam.jp/int/product/dr-10c/spec