Hello People, happy new year all,
I haven't been on diy for so long it's shame. Anyway, I have a few electronics questions for all you people who've serviced tapedecks in the past (or now, but it sure ain't anywhere near where I live: repair have all closed).
1- How does bias affect the recording process? Is it just a manner of rescattering the magnetic particles in a similar fashion before they get rearranged by the record head?
2- How is bias injected into the record signal, by resistive mixing, I've tried to look on the few tapedeck schematics I have, but I can't really figure it out. I remember seeing an LC trap that prevents the bias from going back into the record head amplifier, I understand that.
3- If you were to look at the signal on the record head, what would it look like? bias frequency riding on the audio? what's the ratio of bias freq. to audio freq?
thank you all!
gain-wire
ps: I have looked in the many books aI have, searched the internet, but I have not yet found a source of information I understand. I guess I require interactive help! 😛
I haven't been on diy for so long it's shame. Anyway, I have a few electronics questions for all you people who've serviced tapedecks in the past (or now, but it sure ain't anywhere near where I live: repair have all closed).
1- How does bias affect the recording process? Is it just a manner of rescattering the magnetic particles in a similar fashion before they get rearranged by the record head?
2- How is bias injected into the record signal, by resistive mixing, I've tried to look on the few tapedeck schematics I have, but I can't really figure it out. I remember seeing an LC trap that prevents the bias from going back into the record head amplifier, I understand that.
3- If you were to look at the signal on the record head, what would it look like? bias frequency riding on the audio? what's the ratio of bias freq. to audio freq?
thank you all!
gain-wire
ps: I have looked in the many books aI have, searched the internet, but I have not yet found a source of information I understand. I guess I require interactive help! 😛
I used to love tape as a format but its been a while now since I did any practical alignment.
1. The bias pushes the recording process into the more linear part of the magnetic curve. Thats as clear as mud isn't it 🙂 It's a bit like biasing a transistor or valve to push it into the linear part of the transfer characteristic. Magnetic tape is very non linear at low magnetising levels so the bias overcomes that. It also (I think 😀) puts the recording head into the more linear part of the "B-H" curve,
http://en.wikipedia.org/wiki/Saturation_(magnetic)
Bias can be DC as well as AC. DC bias is "noisy".
2. Its often injected right on top of the record head via a capacitor (that could be adjustable in some models) or via a cap and series pot (the pot sets the bias current). The LC trap you refer to is just to stop the AC bias getting back into the recording amp.
It is just mixed. The head doesn't appear as a short (due to the winding resistance) at high frequency and so the bias signal is often in the tens of volts range although it's actually the "bias current" that matters, the higher currents used for chrome and metal tape formulations. We talk of voltage and measure the "bias voltage" but its the current that's important.
3. The signal on the head is just as you describe, an AC bias (always a sinewave) with any audio modulating it. Common bias frequencies (if I remember right) were 85 Khz and 105Khz. The higher was usually found on top quality decks that had extended response, and they needed the appropriate heads for that.
1. The bias pushes the recording process into the more linear part of the magnetic curve. Thats as clear as mud isn't it 🙂 It's a bit like biasing a transistor or valve to push it into the linear part of the transfer characteristic. Magnetic tape is very non linear at low magnetising levels so the bias overcomes that. It also (I think 😀) puts the recording head into the more linear part of the "B-H" curve,
http://en.wikipedia.org/wiki/Saturation_(magnetic)
Bias can be DC as well as AC. DC bias is "noisy".
2. Its often injected right on top of the record head via a capacitor (that could be adjustable in some models) or via a cap and series pot (the pot sets the bias current). The LC trap you refer to is just to stop the AC bias getting back into the recording amp.
It is just mixed. The head doesn't appear as a short (due to the winding resistance) at high frequency and so the bias signal is often in the tens of volts range although it's actually the "bias current" that matters, the higher currents used for chrome and metal tape formulations. We talk of voltage and measure the "bias voltage" but its the current that's important.
3. The signal on the head is just as you describe, an AC bias (always a sinewave) with any audio modulating it. Common bias frequencies (if I remember right) were 85 Khz and 105Khz. The higher was usually found on top quality decks that had extended response, and they needed the appropriate heads for that.
Thats good.
One bit doesn't read too well though 🙂 The winding resistance of the head 😀
What I meant is that the head, despite having a low DC resistance, has high inductance and so appears as a "high resistance or impedance" at the bias frequency.
One bit doesn't read too well though 🙂 The winding resistance of the head 😀
What I meant is that the head, despite having a low DC resistance, has high inductance and so appears as a "high resistance or impedance" at the bias frequency.
The old rule of thumb was that bias frequency needed to be at least 5 times the highest recorded audio frequency, hence most later professional tape machines would use a frequency in excess of 100 Khz. Since bias frequency is also normally erase frequency this could cause a problem on very old machines with metal erase heads, due to losses. This problem largely disappeared with the advent of ferrite erase heads in the early sixties. Some early machines used a lowish erase frequency and picked off the third harmonic to use for bias.
If you look at the voltage across a head during recording on a scope you will be hard put to see the record signal as it is swamped by the bias. As has been mentioned you need to look at the current.
If you look at the voltage across a head during recording on a scope you will be hard put to see the record signal as it is swamped by the bias. As has been mentioned you need to look at the current.
Ok, well, how do you "look" at the current? Also how can you vary the current going through the head, is this not determined by its resistance?
I guess i'll have to look at my tapedeck or open-reel with my scope and see what happens. thanks again!
This was what I was trying to say above. At AC the heads inductance becomes dominant, not the DC resistance. So its "easy" to develop a high bias voltage across the relatively low resistance head becasue the head "appears" at AC as a high impedance.
How do you measure the current... easy 🙂 You need a small (say 10 ohm) resistor in the earthy end of the head and simply scope the voltage waveform across it to derive the current.
But you don't need to do or to know any of this to set up a deck correctly 🙂
A crude way to look at AC bias is that it is a very high-amplitude AC signal that "shakes the ****" out of the magnetic particles on the tape, both linearizing the resulting audio signal and extending its (high) frequency response. The Bias voltage may be upwards of 50V RMS whereas the audio voltage is only around 2-5 V RMS. Yes the bias frequency should be "much higher" than the expected highest audio frequency (I've seen anywhere from 75 to 250kHz). Also, the signal should be as perfect a sine wave as possible to avoid intermodulating (with?) the audio signal.
The Audio signal from the record amp is typically put thru a "high value" resistor and then mixed with the signal from the bias oscillator before routing it to the record head. And yes there is usually some type of bias trap to minimize the amount of bias voltage "backfed" to the record amp. Some decks do put a low value resistor in series with the record head so you can monitor the bias voltage.
"Setting the bias" typically involves recording a high frequency tone (10khz?) on the tape and monitoring it's output level while adjusting the bias control. You find where the peak amplitude is versus bias level then back off the bias by a certain number of dB - manuals I've seen have a table listing the dB level "backoff" for various tape formulations.
FWIW, Sound Technology did a lot of work with biasing and found that correlating bias level versus the 3rd harmonic distortion of a sine wave (typically 1Khz) was a very good way to set the bias - and incorporated it in their 1500 analyzer. Their test applies a 1Khz tone at a level 10dB above what you want as a reference level - say a 250nw/m tone on a playback test tape - then steps it down a dB at a time while simultaneously measuring the 3rd HD - and displaying the results on a CRT. You can see the 3rd HD drop like a rock and observe/adjust the slope of that curve. Can then also set where you want to "peak" your VU meter - which at one time was 3% HD.
Charles
The Audio signal from the record amp is typically put thru a "high value" resistor and then mixed with the signal from the bias oscillator before routing it to the record head. And yes there is usually some type of bias trap to minimize the amount of bias voltage "backfed" to the record amp. Some decks do put a low value resistor in series with the record head so you can monitor the bias voltage.
"Setting the bias" typically involves recording a high frequency tone (10khz?) on the tape and monitoring it's output level while adjusting the bias control. You find where the peak amplitude is versus bias level then back off the bias by a certain number of dB - manuals I've seen have a table listing the dB level "backoff" for various tape formulations.
FWIW, Sound Technology did a lot of work with biasing and found that correlating bias level versus the 3rd harmonic distortion of a sine wave (typically 1Khz) was a very good way to set the bias - and incorporated it in their 1500 analyzer. Their test applies a 1Khz tone at a level 10dB above what you want as a reference level - say a 250nw/m tone on a playback test tape - then steps it down a dB at a time while simultaneously measuring the 3rd HD - and displaying the results on a CRT. You can see the 3rd HD drop like a rock and observe/adjust the slope of that curve. Can then also set where you want to "peak" your VU meter - which at one time was 3% HD.
Charles
Have a look at the Library section on my website - there is a very useful article about the bias in magnetic recording by Terence O'Kelly .
Cheers
Alex
Cheers
Alex
Wow! That's quite a bit higher than the audio signal... But in the end, audio heads' useable frequency range is after 20 to 20000 Hz, roughly... So I guess they can work at double or triple that, but they are so innefficient at doing so you really need to push them!
that's my guess...
What does it mean to peak the VU meter at 3% harmonic Distortion?
Thanks again!
"What does it mean to peak the VU meter at 3% harmonic Distortion?"
If you think about it, what's a VU meter actually telling you? It's responding, somehow, to the audio level you are trying to record. You probably want it to give you some representation of the full, flat (un-equalized) audio spectrum, and respond as fast as it can, realizing that it has inertia and can't really follow the peaks - that's why some "modern" recorders also included a "Peak Level LED".
That said, as you increase the amount of signal recorded on the tape while looking at the harmonic distortion, once you get to a certain level, the HD increases very rapidly, indicating that you are approaching tape saturation. It makes sense that if you can, you try to set a point, somewhere on the upper scale of your machine's VU meter that correlates to some known level of HD, above which you may not want to go. I think that most VU meters indicate up to (at least) "+3", and my memory is that a lot of studios set the 3% HD level (for a particular tape formulation) to this +3. Just to keep you happy, please understand that once this level is set and the signal backed off so that the VU meter now reads "0", the HD at this level is probably down in the tenths of a percent. In the end it's how it sounds - right??
I'm actually going to be doing some dubbing shortly and have calibrated the meters on my Revox to the above level (for new RMG LPR35 tape). Will listen to the results while varying the record level and attempt to correlate audible distortion to a (peaking) meter deflection - in an attempt to find out that - let's say, I drive the meter to +3 (peaks) and not hear any problem; or maybe I need to back off a bit, or can I maybe I can actually "hit it harder".
In the end it's how it sounds -right?
Charles
If you think about it, what's a VU meter actually telling you? It's responding, somehow, to the audio level you are trying to record. You probably want it to give you some representation of the full, flat (un-equalized) audio spectrum, and respond as fast as it can, realizing that it has inertia and can't really follow the peaks - that's why some "modern" recorders also included a "Peak Level LED".
That said, as you increase the amount of signal recorded on the tape while looking at the harmonic distortion, once you get to a certain level, the HD increases very rapidly, indicating that you are approaching tape saturation. It makes sense that if you can, you try to set a point, somewhere on the upper scale of your machine's VU meter that correlates to some known level of HD, above which you may not want to go. I think that most VU meters indicate up to (at least) "+3", and my memory is that a lot of studios set the 3% HD level (for a particular tape formulation) to this +3. Just to keep you happy, please understand that once this level is set and the signal backed off so that the VU meter now reads "0", the HD at this level is probably down in the tenths of a percent. In the end it's how it sounds - right??
I'm actually going to be doing some dubbing shortly and have calibrated the meters on my Revox to the above level (for new RMG LPR35 tape). Will listen to the results while varying the record level and attempt to correlate audible distortion to a (peaking) meter deflection - in an attempt to find out that - let's say, I drive the meter to +3 (peaks) and not hear any problem; or maybe I need to back off a bit, or can I maybe I can actually "hit it harder".
In the end it's how it sounds -right?
Charles
Last edited:
- Status
- Not open for further replies.