Tascam M-3500 refurbishment

I would contact the seller for an actual e-mail and contact.

I have learned my lesson about storing links that don't work at a later date when you specifically need them!

Always good to have a definite contact.
 
Boy am I glad I found this thread! Really nice job on the cleaning, Tim!

I have an M3700, USA version, no +4 kit, 32 channels and full meterbridge. Couple of things I can clear up about the automation, since it came up: it's VCA, no moving anything, but it will control fader moves, channel mutes, EQ in/out, and aux mutes. It can function in dynamic and snapshot modes, the former sync'd to MTC or SMPTE with dip switches that also dictate FPS. I use dynamic mode for automated mixing and I sometimes use the snapshot just for recalling between sessions so my life doesn't get too knocked off balance every time a studiomate musses up the fader positions while I'm not around (they're all ITB people! :spank: ) Basically, I take a snapshot of all the switch and fader positions so when I come back to the mix to find some faders knocked around and buttons bumped I can load that snapshot, put the computer in Null Mode and the switches will set themselves where they should be and arrows next to the channel faders point to where they should be moved. The global auto modes are typical Read, Write, Update, and Null (the latter is a handy touch both in dynamic and snapshot modes), and you can also put individual channels in and out of read/write with a button next to each fader. The automation on the M3700 is IMHO actually pretty pleasant to use for what it is and was very easy to learn. Only complaint is that the VCAs add noise to the path and they also dull the high end a little bit--but the noise that automated mutes and fade downs prevents more than makes up for the noise added by the VCAs. I'm just glad Tascam made the system bypassable so channels you're not automating are not affected!

Anyway, my M3700 is fairly clean (previously lived in a church from what I know) and it's fully functional and free of scratchiness, but it's fairly noisy and most likely hasn't been recapped. I had the PSU serviced when I bought it but I don't think its caps were part of the repair--I'd have to check the invoice.

Any thoughts on what can be done to quiet this guy down? Eventually I will trade this puppy up for a large format console so I don't want to put more cash into it than I have to, but for now I'm using it and would rather not sell it before lowering the noise floor anyway, though I won't be going for the whole cap upgrading, re-chipping and so-on. Are there certain caps on the channels (even just 1-16, my most-used) and in the master that would address the noise, without having to replace all of them? I reckon I'd be smart to start with the PSU caps and see what that does?
 
Last edited:
I remember reading elsewhere on the web that the noise issues on the 3700 could be helped quite a bit by addressing the AC feed to the studio, being careful about not using certain types of lighting like CFL's and having a very clean, filtered power conditioner to take any noise out of the mains that feed the board.

I don't believe the caps in that model are old enough to worry about changing them out.

Cheers! :)
 
The Ghost of FM said:
I remember reading elsewhere on the web that the noise issues on the 3700 could be helped quite a bit by addressing the AC feed to the studio, being careful about not using certain types of lighting like CFL's and having a very clean, filtered power conditioner to take any noise out of the mains that feed the board.

I don't believe the caps in that model are old enough to worry about changing them out.

Cheers! :)
Click to expand...

Thanks! The cap issue is encouraging to know...as for the power...I'm running it on a Furman AR1215 power regulator which is further conditioned by a Blue Circle Noisehound...no fluorescent lights in the studio, though there's plenty of audio gear on the circuits...also I'm only line mixing with this console, but using the balanced mic pres with pads in for balanced line level per the manual. I also boost the aux outs to +4 with Ebtechs, but still more noise as each channel gets un-muted, and especially in the master buss.
 
briank said:
Thanks! The cap issue is encouraging to know...as for the power...I'm running it on a Furman AR1215 power regulator which is further conditioned by a Blue Circle Noisehound...no fluorescent lights in the studio, though there's plenty of audio gear on the circuits...also I'm only line mixing with this console, but using the balanced mic pres with pads in for balanced line level per the manual. I also boost the aux outs to +4 with Ebtechs, but still more noise as each channel gets un-muted, and especially in the master buss.
Click to expand...

Well, if you've done all of that and are still having issues, I'm not sure what else to suggest short of modding the board, which you said you'd rather not do.

Though you might get some small additional noise reduction benefits by looking into star grounding the AC which basically involves having the mixer's ground as the closest wired ground to the room's supply and feeding everything else in the control room from that one ground point/power outlet in the room.

http://www.engineeringharmonics.com/archive-papers/ground_systems.pdf?q=papers/ground_systems.pdf

Cheers! :)
 
You would do well to replace all the lytics in the P/S, increase their capacitance by about 50%, maybe go to higher voltage ratings, and then by-pass each on the board underside with a .01 polyprop or polystyrene cap.

Lytics age poorly and make a lot of noise, and 70's - 80's lytics were not as well made as present makes.

Nichicon PW series are best for the P/S.

Do this and you should notice a huge difference.
 
Greg...That's a 90's mixer we're talking about, not 70's or 80's...and even in my early 80's Tascam gear that I've recapped I have yet to find a cap that has drifted out of spec on value. No, I haven't tested them all but I've tested a good many of them. ESR spec may be a different story and it is unknown to me as I lack an ESR meter, but I concur to not just jump and throwing new caps at the thing.

And regarding grounding...my understanding is that grounding issues will propogate line voltage noise (60-cycle hum) and also open the door to environmental radio frequency interference...we aren't talking about broadband audio noise like I'm assuming you are experiencing.

First thing I would do is measure the noise level. Do you have a true RMS voltmeter? If so I'd get 32 150ohm resistors (cheap), stuff them in between pins 2 and 3 of every mic input, source each channel to mic, defeat the eq section, route each channel to the main buss (and ONLY the main buss), turn the trim pots to about 3:00, raise the channel faders to unity, raise the main faders to unity, turn on your true RMS meter to AC volts and measure the voltage level at the unbalanced main output and post it here...convert it db's if you like or we can do that here after we have AC volts RMS value. This will be a measure of the mixer's nominal self noise at unity gain for 32 channels through the mic pre's at a reasonable real-world gain level. Maybe the mixer is already at spec...maybe not.

At any rate, I don't find that Teac made gross errors or undervalued filter caps in their PSU's, and if there is a noise issue, like I said, you're not going to necessarily rectify it by throwing more filtration at it...and I have yet to find a cap in a Teac unit that is undervalued in voltage rating.

So if the noise level is off spec or you just want to try and improve the stock performance, you certainly *could* replace the filter caps with good quality caps (and I do like the Nichicon PW caps for power supply applications)...if space allows you're not likely to HURT anything by increasing the capacitance value but don"t expect a silver bullet...and the film bypass...well that also is not a one size fits all solution either in my understanding...I believe that a lot of broadband audio noise issues are often related to thermal noise...the wellspring is in the PSU in the bridge rectifiers and regulators. Look at those components in the audio power rails...get online and pull the spec sheets for the stock components and look at the noise specs and then see if you can find replacements with better noise performance...if you are replacing any of those items then make sure the cooling conventions in the PSU appear logical and effective, and get new insulators with the regulators while you're at it if you replace the regulators...get the silica type, not the rubber type. Use high-grade thermal grease between components and any heatsinks.

Now, if you decide you wanna just shotgun all the 'lytics in the PSU I won't try to stop you...I had wonderful results doing so with my M-520 which was really really noisy...I didn't measure anything so I don't know how much the recap did, but it was an unmistakeable improvement...however I wouldn't worry about messing with the mixer caps.

I recapped the PSU for my Soundtracs MX 32 x 8 desk along with regulator, rectifier and heatsink mods/upgrades and it dropped the noise floor by 4 db's...that was worth it. I will say though the heat management in the Soundtracs was, IMO, totally inadequate in stock form and the Teac mixer PSU's I've been able to see inside of are on MUCH better ground stock in terms of heat management than the stock Soundtracs.

So there are some more options and things to consider.

In summary if it were ME I'd be recapping the PSU with Nichicon PW's using the stock values and checking into possible improvements by upgrading the rectifiers and regulators...all AFTER I got a baseline measurement of the noise so I could repeat later and quantify improvements.

You might also consider downloading something like True RTA...its free 1 octave spectrum analyzer software (and full of lots of other handy features) from True Audio (paid versions have higher resolution...1/3 octave...1/6 octave, etc.)...it'll at least give you some quantifiable feedback as to whether or not your issue is broadband noise or something else.
 
Very good points, Cory, to measure a noise floor as that is a good starting point.

Another possibility that I have frequently experienced in high end audio is ground looping and that can cause a ton of noise.

Some Bryston amps have a switch to "lift" ground to rid noise and hum.

Suggestion: make sure you are on a dedicated circuit for all the equipment power feed. Additional connections on a line can add noise by inductance.

Then connect all your equipment to that one source and use something like a Panamax model 5300 or 5400 line conditioner (http://www.panamax.com/products/a-v-components/m5300-pm.aspx) to make sure you are use clean power isolated from the source.

These units use isolation transformers so you are dealing with very clean power.

Worth a try.
 
I agree that it becomes absurd, but when you need that last part that no one else has then there are few options.

At least it's a possible alternative.
 
Thanks, I appreciate all the input very much and hope it all helps other M3x00 users browsing this thread too. I'm an experienced audio engineer but certainly no EE...some of this admittedly goes over my head but very grateful for all the info! I can solder and use a multimeter, though haven't quite figured out my 'scope yet :o

To clarify, yes we're talking about "white" noise floor--I am not experiencing 60 cycle ground hum as such.

Looks like the most sensible approach is to start with the AC delivery in the studio...this is a rented facility and although some noise is due to inductance, for now my hand are fairly tied as far as being able to get this desk onto a dedicated circuit or tweak the grounding scheme--best I can do is give it regulated, conditioned power off a somewhat clogged grid...but I'll do what I can to minimize these issues. Sounds like the next best active thing I can do is tweak the PSU--in my case this is do-able and much preferable to having to pull the quad channel modules apart! I can say, though, I don't have a ton of reason to suspect this console is particularly out of spec. From what I can tell, it led a light to moderate duty life and was sufficiently maintained. Most issue I suspect are due to age, design, and/or physical setup/ergonomics, the latter being the easiest to deal with. That said, the noise floor I'm experiencing is beyond the limits of modern commercial acceptability...
 
"white noise" may be misleading--it's your good old fashioned hisssssssssssssssssssssssssssss noise floor. Definitely the console. With the trims down, the EQs out, and faders and masters up at 0 I do get a higher noise floor than I'd like. Not surprisingly, bring a 16-track project with a few sends and returns to unity and it's a much higher noise floor than I'd like. Almost certainly a compound problem, the more I look into it.
 
I have had a m3700 for many years and I don't think they are any more noisy than most other consoles (except if we are talking mega price). What I have found is that if you mute all the effects sends and the tape monitor faders (even if they are turned completely off) the noise floor drops heaps. Then only un-mute the sends and tape returns you are using.

The other thing to watch out for is grounding between the console and the recording device and between the console and the monitor amps. I run a big fat ground with between the console (fixed to the chassis), amp chassis and recorder chassis.

The other thing I did that reduced noise was to put isolation transformers on the stereo lines in and out that connect to the computer, this reduced the digital ground buzz that comes via the firewire (or USB).

I also have an isolation transformer on the mains supply due to buzzes and noises that come in on the mains.

One thing I was never sure of was that this console is a 240 volt 50 hz in Australia, it could be that the US 110 volt 60 hz consoles had more problems.

The M3700 (at lest the later ones like mine) had improved earthing over the M3500 and so some of the modes recommended for the 3500 are not needed on the 3700.

Cheers
Alan.
 
Briank, I really think it would be valuable to conduct the test I mentioned in my last post to get a baseline of what kind of noise you are experiencing (the one where you stuff the tails of a 150ohm resistor between pins 2 and 3 of each mic input blahblahblah....)

And I want to emphasize that, unless I'm wrong, hissssssssssssss is not typically the kind of "noise" you're going to get from grounding problems.

If you are capable with a soldering iron, a PSU recap is a fairly quick job and we'll be happy to help with any sticky bits you come up against. If you don't want to deal with "should I change this value" or "do I tweak that component", just swap out the old caps for new ones of the same spec. I realize that there is value to making changes, but I've also come to peace with the fact that neither am I an EE...I don't know how to truly evaluate what could/should be changed nor do I know how to quantitatively test the results...based upon my own experience and knowledge as well as the comments of much more knowledgeable folks whom I trust, Teac engineers knew what they were doing. Their products aren't "boutique" but are well engineered and designed, and are built for quality and reliability and sound good, especially when considering the price point. And the innovations abound when you look at the history of their product lines...many "firsts". So I don't feel in any way that dropping in components of stock values is anything to feel regret over. So that's just something to think about.

Again, I did a recap on my M-520 PSU...I wrote about it in my "Tascam M-520 Story..." thread...there was a very unexpected drop in the noise floor afterward. Bad caps typically cause pops and crackles in the audio path, not hisssssssss...but hey, I know what I experienced. I'm sure a aged filter cap in a PSU could throw a regulator for a loop...no way for me to know why the recap did what it did, but it did. Worth a shot but try and measure the noise before and after.

Again, I comfortably recommend the Nichicon PW series for PS caps...why? Nothing shocking...Mouser stocks them, they are very affordable but (IMO) their quality control is great, I've never had a problem with one and their spec is great for the price. A complete PSU recap will cost you around $20 including shipping the parts.

And since we're talking about grounding, I totally don't want to step on Alan's toes...he's got something that is working for him but you want to be careful when you strap chassis together...if there is some other established ground path you will create a loop which is an open door to the noisy crap pungent in any environment. Whenever possible you want to have one ground source for all your audio equipment, and ideally that one source is the wall receptacle. Google something like "Neil Muncy AES grounding" or something like that. There's a pdf floating around of a paper he wrote for the AES on audio grounding and it is a really great and valuable read...covers the dysfunctionality between manufacturers in grounding schemes and how to identify and resolve problems in your own studio. Its a real problem...you have signal grounding and chassis grounding and the industry can't agree on a standard. This is why I personally wouldn't be strapping equipment together, but tying it all back to the ground source in the wall receptacle and THEN start analyzing the signal and chassis grounding schemes for the different units in the signal chain if I'm having problems.
 
Last edited:
With regard to Sweetbeats,

I should have stated that you only bond the chassis if an improvement happens, and I am talking about the dreded 50 hz (60 hz USA) hum. if no improvment don't do it.

The reason that I understand that chassis bonding can work is that the loop will travel via the least resistance, so if it's got a fat cable to travel down it reduces on the audio cable, and this sometimes seems to be the case, but it could be all bull. Also the bonding in no way means you can disconnect the mains earth, DO NOT Disconnect mains earths.

Earth loops are the bane of studios and sometimes very difficault to get rid of.

Cheers
Alan.
 
witzendoz said:
The reason that I understand that chassis bonding can work is that the loop will travel via the least resistance, so if it's got a fat cable to travel down it reduces on the audio cable, and this sometimes seems to be the case, but it could be all bull. Also the bonding in no way means you can disconnect the mains earth, DO NOT Disconnect mains earths.

Earth loops are the bane of studios and sometimes very difficault to get rid of.
Click to expand...

Yeah...that's why I tried to make the disclaimer...if its working for you then there's not really any point in me discussing or criticizing what you are doing.

You are correct about the path of least resistance but the other issue at hand is that by putting a secondary earthing path in place you aren't eliminating the inital path if there is one. At some point signal ground couples to chassis ground or at least it should, and if it doesn't that's the key problem to be rectified...and if it DOES couple to the chassis and you put that secondary chassis to chassis strap in place then the original signal-ground-to-chassis connection becomes in open inductor for environmental magnetic and RF noise that won't be shunted.

Here's my application of this: I have a dedicated circuit to power the audio equipment in the studio that I can confirm ties to the physical earth at the service entrance to the building. So that takes care of a single path to physical earth ground. The dedicated "home run" for the audio gear just minimizes issues of line noise from motors and such. Power conditioners connect to that one circuit at the same wall receptacle which carries all ground connections to that single point. This is proper star grounding. My problem to rectify (if there is audible garbage to deal with) is that my Soundtracs mixer has non-ideal audio grounding...all the mic inputs, for example, have a pin 1 problem: pin 1 DOES tie to the chassis but they should tie to the chassis as close as possible to the physical junction. That means that pin 1 should physically connect to the chassis right there at the jack. Soundtracs star-grounded ALL the signal grounds to several points (like in groups of 8 channels) inside the mixer, so there is maybe 10 or 20 FEET of basically inductor cabling inside the mixer. Now, the chassis is all metal so it probably provides decent shielding from environmental magnetic and RF noise, but what about inside the mixer? That depends on how the wires are run and proximity to the "guts". So once I get the studio setup I'll just have to see but since everything audio will be going through my mixer I will likely mod the pin 1 connections at some point and connect them to the chassis at each respective jack. Then as other issues crop up I will analyze how signal grounds were handled in other pieces of gear, and if certain pieces of gear have a 2-prong power plug then I will modify appropriately so they tie to the chassis ground present where they source power (the power conditioners which have a nice fat lug on them for that very purpose).

If you can get a hold of Niel Muncy's 1995 AES publication on the pin 1 problem I highly recommend it. It really is a great and valuable read. He's a really great guy...We've had a couple good chats over this grounding stuff. He's made countless contributions to the audio industry.
 
You guys have given me some food for thought and ideas to play with--thanks again very much! I'm going to do a little more troubleshooting, play around with power and setup, try the "noise test" and make sure I'm barking up the right tree (or at least the right forest) and probably look into the PSU caps especially since it's a pretty cheap and easy project.

Incidentally, I just noticed the mic gain pot on channel 3 seems to be a little erratic and cleaning didn't help--any suggestions on sourcing a replacement?
 
Last edited:
You must log in or register to reply here.
Back
Top