Tascam 58-OB Story...

[sweetbeats]

input cal done with oscillator loaded or unloaded?

I should just look back on my own thread...I think I resolved this a long time ago...

When I am setting the input calibration, am I correct that I want to set my oscillator output level with it connect (loaded) with the deck input(s)?

1. Connect oscillator output to atr input.
2. Dial in my reference voltage level using my AF level meter that is connected in parallel to the (loaded) oscillator output.
3. Disconnect the AF level meter from the oscillator output and connect it to the respective atr output and tweaker the input level trimmer on the atr to get the same reference voltage coming out of the deck that I just measured going into the deck.

Is that right? I'm thinking it is because the whole point of calibrating the input level is so that your audio path reads a consistent reference level throughout, and I certainly wouldn't be trying to set levels on program material with it disconnected from the deck ...that just wouldn't work, so program peaks are going to be set with the source connected to the deck, so I should do the same when setting the oscillator output right??

Seems like a really silly question, but one of those things that happens...anybody else ever write a word and then realize it looks funny and then you start to ask yourself if it is really a word? Maybe that's just me...Anyway, this input level setting question is kinda like that...

 

[sweetbeats]

Stoopid tensions!!!!

In the middle of the record calibration my tension adjustments seem to be drifting!!

The reels creep in EDIT, and take forever to come to a stop in REW.

*@#%*^&)@!)#%$(

Top cover off...tweaker on trimmers.

ARG!

 

[sweetbeats]

Could it be heat-related?

The "studio" is pretty warm. I've got an AC unit but it is pretty ineffective with the lights, computer and all the other rack gear on, not to mention the 58...MAN it puts out some heat...I haven't had it on this long in the warm weather like this, plus its the first time its been on for any length of time with its skin on...

 

[sweetbeats]

Bias level wierdness

This goes back a ways...I noticed this and commented on it in a post awhile ago.

When I'm setting the bias level, I'm supposed to be able to turn the bias level trimmer all the way counter-clockwise while recording a 10kHz tone and monitoring with the repro head, and then as I turn it clockwise the level is supposed to crest and then drop. Then I'm supposed to find that crest level and turn the trimmer counter-clockwise again so that the repro level is -4 ~ -5dB below that crest level...my levels never crest.

My bias level trimmers hit the clockwise stop before I reach a level crest or at least as far as I can tell.

I originally thought it might have something to do with where previous ownership had set the repro hi eq trims because some setting were definitely tweakered, so I started experimenting with those a bit, and with my bias level set as well as I can set them, when I am recording a 20kHz tone and monitoring with the repro head my eq trimmers crest the reproduced tone! In other words the repro level of the 20kHz tone decreases as I turn the trimmer counter-clockwise. If I then turn it clockwise the level crests and then drops as I continue to the clockwise stop of the trimmmer. Wha?

BTW, I am getting reproduce frequency response of 0dB ~ +1dB from 40Hz to 20kHz using 40Hz, 100Hz, 1kHz, 10kHz and 20kHz tones using RMGI SM911, and the needles are very steady across all tracks, so that is good...

There is a video of this debacle here. It is about 26Mb in size.

Help?

 

[evm1024]

Setting bias

OK so my wife thinks I'm weird because as I was reading your post I was turning my (Imaginary) adjuster with my hand in the air.

Rather than respond directly to your post I'll just ramble on.....

With the bias at minimum (all the way counter clockwise) there is very little bias "injected" to the recording heads. The repro heads show very little signal and what signal there is is quite distorted.

Turning the bias adjustment clockwise (increases the capacitance of the adjustment cap) and couples more bias signal to the record heads. The repro head signal levels starts going up and the distortion starts going down.

At some point as you continue to turn the adjustment clockwise (increasing the bias level) the repro head will reach a peak level (showing the maximum signal level transfer to tape). This is not the lowest distortion level.

Continue to increase the bias level (by turning more clockwise) and you will see the repro level decrease. When you have increased the bias level such that the repro level is 4 dB lower than the peak level you saw (4 dB overbias) you will be at the optimal bias level.

Of course 4 dB overbias is just a typical number. Each tape/head conbination has its own optimal overbias.

The key here is to continue to add more bias past the peak signal to the lowest distortion point.


You can set this another way. Record a 7 Hz signal starting with the bias control fully minimum (counterclockwise). Listen to the signal through the repro head. What you hear is the distortion products (unless your ears are much better than mine )

Now continue listening and increase the bias. You will at some point hear the distortion go away then return as you continue to increase the bias. The quietest pint is the optimal bias setting and the point with the lowest distortion.

You can then record a 16 kHz (or whatever your bias adjust frequence is speced at) at a known level then back off the bias level to find the peak and calculate the optimal overbias for this tape/head combo.

Regards, Ethan

 

[sweetbeats]

Confound it all, Ethan...

Continue to increase the bias level (by turning more clockwise) and you will see the repro level decrease.

That last part is what is not happening. The repro level never decreases.

This is not good is it?

 

[evm1024]

That last part is what is not happening. The repro level never decreases.

This is not good is it?

Keep tweaking. There is some interaction between it all.

You might want to get a scope on the bias osc and amps to verify that they are working correctly (have enough level and right freq range).

In general not good.

Give the 7 Hz test a try to see if there is a min distortion point.

Also, do the specs say to use 20 kHz? I'm thinking that a lower test tone is generally used but then again i've never called a 58.

--Ethan

 

[sweetbeats]

Keep tweaking. There is some interaction between it all.

okay...I will...

You might want to get a scope on the bias osc and amps to verify that they are working correctly (have enough level and right freq range).

I don't have an analog scope, just a PC-based scope, but it is functional. The bias oscillator and bias amps are found on each channel amp card right? They look kinda like relays but they have a trimmer on them right? Any idea how to find the level and frequency spec? I'll look in the manual, but...

In general not good.

What could it mean?

Also, do the specs say to use 20 kHz?

No, no. 10kHz is the specified frequency for setting the bias. I was using the 20kHz tone to set the hi-freq response...it was a bit low on track 8 and so I went to tweak it and the level DROPPED when I turned the hi-eq trimmer clockwise. That's how I discovered the strangeness with the 20kHz tone but that didn't have anything directly to do with my bias level concern. When setting (or attempting to set) the bias level I am using the 10kHz tone, but the repro level never drops...it is still rising when the bias level trimmer hits the clockwise stop. They (the hi-eq trimmer effect at 20kHz, and the bias level issue) are separate issues but I am hoping they are related.

Ethan, did you happen to watch the video? I'm not being too clear in these posts and the video may help to clarify.

I'm scared...and unhappy...at the moment.

Ethan, thank you for helping.

 

[evm1024]

Late at night

Never got the chance to watch the video.... Just got out of a hell month at work (actually not hell just a lot of work and no time to think of anything else)
We have a research cruise going on and I may vanish for a day or 2 when there are problems.

I would say to forget about the 20 kHz levels as a diagonstic. Get what you can there but keep the (up to 12 kHz) lower highs as flat as you can.

Roll off at 20 kHz could be aging (those old caps) components, wear on heads or some other such thing. That area of the spectrum is not heard directly but plays an important roll in the harmonics that are present. Bring 3,4,5 dB down is nothing IMHO. Just being there is what is important.

Oh, and not that many mics or instrument go there.

Remember your deck is not a scientific instrument used to capture and measure sounds. It is a musical instrument and has 'color" of its own.

I kinda doubt that your PC scope will work (but then again I don't have much experience with them) as that the bias osc is up in the 145 kHz range and I presume that your audio card has a 20 kHz limit.

I've got a Tek 465 that you can use when ever you want (100 MHz analog scope).

Bias level and freq should be in the book. Generally it is not messed with much. You would be looking for differences between channels.

Not knowing the 58....The master osc sets the freq and each channel card has its own bias amp to set the level and perhaps enable (on/off). Yes they are in the big cans with the adjustment.

How does your deck sound?

_Ethan

 

[sweetbeats]

Later at night...

I am shocked and excited that my little story is nearly about to surpass 4000 views...and only 3500 of them are me!

Ethan, thank you again for your postings...

I would say to forget about the 20 kHz levels as a diagonstic.

The only reason I'm doing anything with 20kHz is because the manual says to review playback and record response and tweak to maintain +/-3dB from 40-20000Hz. I will stop trying to get it to 0 at 20kHz though.

Remember your deck is not a scientific instrument used to capture and measure sounds. It is a musical instrument and has 'color" of its own.

Yes, yes. Thank you. You and cjacek have a way of bringing me back reality. You guys are like...Cory whisperers...

Bias level and freq should be in the book. Generally it is not messed with much. You would be looking for differences between channels.

Not knowing the 58....The master osc sets the freq and each channel card has its own bias amp to set the level and perhaps enable (on/off). Yes they are in the big cans with the adjustment.

I did some (re)reading. Yes, each amp card has a module for bias tuning, a master oscillator (which confuses me because there is a separate card with a master oscillator as well), and two traps.

I did some experimenting this evening...

The bias tuning is measured at two testpoints on the amp card, and with all the record enable switches set to ON, and the deck in record mode monitoring the repro head, and with the bias level trimmer "set correctly to the 3 o'clock position" you are supposed to tweak the adjustment screw in the level module until the reading is at 0.35VDC using a meter with "less than 100pF floating resistance"...I have no idea if my meter meets the specifications...it is a Fluke 85 and the manual says nothing of the floating capacitance......plus I have no idea in what position the level trimmer is pointing to 3 o'clock. Is that looking at the end of the card in horizontal or vertical position, and what is the reference pointer (the hour hand if you will) on the trimmer???

Regardless, I couldn't help trying some things...the voltage at the testpoints was indeed about 0.35VDC, and again, at that setting, clockwise rotation of the bias level trimmer never results in a peak and then fall on the VU meter of the 10kHz tone...only peaking until the trimmer hits the clockwise stop. Now, if I raise the voltage on the bias tuning module to 0.50VDC, now the bias level trimmer works like its supposed to! I turn it full CCW, slowly turn it CW and after a bit the VU presents a peak and then the level rolls off with continued CW rotation.

So, that was all I had time for tonight...it works, but now I've got to try and figure out why I have to set the voltage of the tuning module higher to get it to work right...and it is going to be a pain trying to figure out a baseline for the repro hi-eq trimmer...it is whacked and since it has an effect on the bias setting...I'm not sure where to start.

But at least there is some good news here.

How does your deck sound?

Haven't even recorded program material yet , but the tones are clean...in fact the 10kHz tone sounds clean regardless of the bias level setting.

 

[sweetbeats]

Wondering if it could be bad/failing caps that could be causing tension issues...

Just a thought...It is about 25 years old, and I think it sat unused for a number of years...

 

[sweetbeats]

Hm...

So I did what I usually do when I reach a complete wall in this story...I call Tascam and talk to Jim or Jimmy. I got Jimmy today.

I updated him on the tape tension issue. He agreed that it is possible that there could be a failing capacitor somewhere in the deck that could cause trouble, but in the same breath he said "it could be a whole bunch o' things...). He suggested that it may be a bad reel motor. I've brought that up in the past in this thread I think...the supply motor, which is where the problems exist, is indeed a replacement. I think I actually pulled it from my 48. It runs nice, strong and quiet, and it is the appropriate part number, but Jimmy confirmed that, without some pretty sophisticated test equipment, there really is no way to test the motor and confirm yay or nay that it is under spec or otherwise a contributor to the problem. I suggested that I could just swap reel motors and see if the problem follows. He liked that idea, and I do too. It would be a little bit of a job and then of course all the tension stuff would need to be reset, but at least then I'd know if that is the problem and then I could move forward on a solution. So we'll see if I get into that.

I think pretty soon here I going to need to make a decision as to whether or not I record my patiently awaiting client using the 58, or just go digital and get it done. Maybe I'd think more clearly and enjoy the journey more if I weren't under pressure...and I want to enjoy it. I love to learn and tinker, and at the end have an instrument that I know inside and out as much as I am capable...at least as far as the adjustments and setup goes...to reach a point of maintenance efficiency and confidence such that production downtime is little, and the simplicity of atr production, that absence of digital bells and whistles and presence of basic tactile hardware controls, can be fully experienced.

Yes I could just let it go as I indicated a few posts ago and move on, but some stuff is bugging me...the REW tape packs are pretty nasty and I don't like that. I'm probably being too picky, but its hard not to be concerned when occasionally the tape rubs on playback because of a rogue section of wraps, and I run all my tape onto the supply reel (from tails-out storage) so at the start of anything I'm rewinding the full tape and its just not what it should be. PLAY, FWD or FWD SPOOL packs are beautiful. The other thing though that is more concerning and has driven my mind back to tackling the problem rather than ignoring it is that when I'm in REW SPOOL and I hit STOP or PLAY the back tension can't keep up and the supply tension arm bottoms out hard and the tape just gets jerked. I actually have to go from REW SPOOL to REW just for a split second and then enter the next transport command just to be gentle on the tape.

Okay...so back to the phone call: I then moved on to the bias tuning issue. My first question was about the stated requirement in the manual that an AC level meter with a "floating capacitance of less than 100pF" be used. I can't even figure out for certain what floating capacitance is...not sure if it is the same or similar to self-capacitance, and in any case I can't determine whether or not my Fluke 85 meets that spec. Jimmy doesn't know the answer to either of those questions either...I feel better about that I guess. He transferred me to Jim Finch. Had to leave a voicemail.

So anyway, with last night's experiments I know that the bias circuitry is functioning correctly, but it seems I have to force certain inappropriate settings to get it to perform as it should, and that is hopefully what Jim Finch can help me nail down.

 

[pianodano]

Haven't even recorded program material yet , but the tones are clean...in fact the 10kHz tone sounds clean regardless of the bias level setting.

No doubt that will. Record 40 cycles and listen for rocks. (no kidding) That will tell the tale.

 

[sweetbeats]

Danny, you mean while tweaking the bias level?

 

[pianodano]

Danny, you mean while tweaking the bias level?

Cory,


Yes I do. Trying to get 0db at 10 or 20K is probably going to saturate the tape.

Certainly you should do a normal bias. But as a test, try recording that low frequency tone while adjusting bias should quickly expose the rocks. Minimize them. That's what the old timers called them because that's what it will sound like. Try it on a channel or 2.


Just my 2c.

 

[analogkid]

I'm not sure how big a deal it is to have to tweak the bias voltage slightly - this is analog equipment as someone else pointed out. The real question is can you tweak the deck so that tracks can be recorded that sound good and can be mixed down later in a useful fashion? Everything else is really just icing.



AK

 

[pianodano]

I'm not sure how big a deal it is to have to tweak the bias voltage slightly - this is analog equipment as someone else pointed out. The real question is can you tweak the deck so that tracks can be recorded that sound good and can be mixed down later in a useful fashion? Everything else is really just icing.



AK

You are right on the money. Tascam, even when ideally setup with the warehouse full of test gear that they spoke of having, will have -3 to +3 deviation on their field setup graph. The important thing here is - "how does it sound" ?

A thread is going on now over at GS. http://www.gearslutz.com/board/so-much-gear-so-little-time/249762-biasing-techniques.html

 

[sweetbeats]

You are right on the money. Tascam, even when ideally setup with the warehouse full of test gear that they spoke of having, will have -3 to +3 deviation on their field setup graph. The important thing here is - "how does it sound" ?

Yeah guys, but you are talking to a guy that doesn't know how its supposed to sound. That's the tough spot for me. Practically all of you who have helped and contributed to my journey have actually worked with an atr! Many of you, respectfully, have a ton of experience with them. For all intents and purposes my ears and skills have developed almost entirely working with the digital domain. I can't just slap tape on, roll it and go "that sounds right/good". I'm trying to get the deck into spec so I have a starting point...I need to get my sea legs so to speak but I can't do that unless the ocean is rocking the way it is supposed to rock. This is all moonlighting stuff for me...nobody else in convenient proximity to me has any clue what I'm doing...anybody I do know is working with digital...I have no analog recorder/reproducer to refer to.

This bias thing is a major deal in my mind...the way the deck is at present, with the bias tuning module voltage set so that there isn't enough bias level available at the trimmer to reduce the repro level of the 10kHz tone...isn't that important to resolve?? I mean, if I'm understanding this correctly, that means that the deck is seriously underbiased and its gonna sound dirty.

I'm not sure how big a deal it is to have to tweak the bias voltage slightly - this is analog equipment as someone else pointed out. The real question is can you tweak the deck so that tracks can be recorded that sound good and can be mixed down later in a useful fashion? Everything else is really just icing.

AK, this is exactly what I'm talking about...you're supposing it may not be a big deal, but introducing bias into the signal path to tape is an essential element of analog tape recording... I can't just turn the trimmer, see that there isn't enough bias to effect the signal and go "Well, its analog...it would be icing to have the bias circuit effecting the response, but oh well..." That "slight" bias tuning voltage adjustment is the difference between the bias circuit doing what what it is supposed to do or not, period. Slight or not the outcome is significant.

Yes, I'll listen for rocks.

Yes I will continue to not worry about whether or not the response is flat, but let me clarify that I haven't been tweaking to get it flat, that is the response curve that the deck is presenting during this first calibration where I have a little more comfort/experience with the deck.

I do understand that the repro curve is expected to have a +/-3dB swing.

My deck is doing something different and I'm guessing it is because it was not setup right by the last guy. I want to fix that.

Even the input response is expected to have a +/-2dB deviation. My deck isn't doing that...I want to fix that.

What should I expect for input response at:

40Hz
100Hz
1kHz
10kHz
20kHz



What should I expect for repro response at:

40Hz
100Hz
1kHz
10kHz
20kHz


And I am using those frequencies because those are the frequencies that the manual specifies are to be referenced when setting the input and repro response.

Guys, I know I've spent a ton of time and put up a ton of posts that are all about dickering with this thing. You are all probably getting tired of the questions and the tweaking and the tweaking and the questions...You are all probably wondering "Goodness when is this guy going to put his tools away and make some music???!?"

Believe it or not, I am right there with you, but my respect and appreciation for the gear that I've crossed paths with that is in my possession is beyond my dreams...never thought I could own anything like the 58...yeah its not what the snobs call pro, but it is way more pro than I ever thought would be in my reach, so its pro-pro-pro to me and I want it to last a long time. Frankly I'm scared of ruining it by running it in production if something is out of whack. That means setting it up so that it runs the way its supposed to. I can tell that it has not been setup right mechanically or electronically, so before it goes into production I want to get it back to a baseline of sorts.

And to answer the question "How does it sound?

Danny, I don't know, man. I don't know, because I don't know what I'm supposed to hear...what I'm listening for, so all I can do is look at the needles and the screen on my DMM until I see it behaving the way the manual says its supposed to and then mic up the kit, plug in the bass and see if my heart and gut were right...

This is a rant, and I'm sorry about that. I truly appreciate the help and advice. I mean no disrespect and I am certain that everything you all post is with good intent, but I feel that there is some misunderstanding, either you of my logic, or me of my logic.

If I'm nuts, tell me...please.

Is it wrong to try and get the deck into spec???

 

[sweetbeats]

I need to get the response back to some sort of baseline.

If I pull bias out of the picture, what should I see for input response at:

40Hz
100Hz
1kHz
10kHz
20kHz

What should I expect for repro response at:

40Hz
100Hz
1kHz
10kHz
20kHz

This would help because I'm concerned that in my inexperience I have adjusted eq trimmers based on grossly misadjusted bias leevel trimmers from past ownership...so now the eq gets whacked and now I'm chasing it with bias...one begets the other and I can't find the path anymore.

 

[evm1024]

You want response curves? Here you go.

Here are some graphs.... http://www.endino.com/graphs/

You will find a couple of Ztascam machines there which would illistrate how the response curve of your 58 could look. Notice the MS-16 high end.... Most of this is due to the heads and not to the channel cards.