New Posts New Posts RSS Feed - 1970s Design Indulgence
  FAQ FAQ  Forum Search   Events   Register Register  Login Login

Welcome to the Graham Slee Audio Products Owners Club

 

Open to all owners plus those contemplating the purchase of a Graham Slee audio product wishing to use our loaner program: join here
Subscribe to our newsletter here (Rules on posting can be found here)

1970s Design Indulgence
 Post Reply Post Reply Page  <1 255256257258259 281>
Author
  Topic Search Topic Search  Topic Options Topic Options
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Online
Points: 15025 		Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 28 Nov 2022 at 10:12am
This is the 7.5 ips frequency response. Like RIAA for vinyl, it is an equalisation response, and if you read off the 1kHz level, it's 45dB, and then you go to 50Hz, and it doesn't quite reach 77.5dB, but it's close.

ERROR: 1kHz level is 52.5dB 

7.5 ips tape EQ frequency response plot

Then if you turned that upside down, like a mirror image or reflection in a lake, the top of the mountain becomes the bottom, and 77.5 minus 45 is 32.5, and -32.5dB is 1/42.

So if at 1kHz the tape head outputs 1mV, at 50Hz it outputs 1/42nd mV, which is 24uV.

24uV

And this curve is amplifying that -32.5dB by nearly +32.5dB so that the tape head and tape equaliser combined responses, give a flat frequency response.

And if you remember what was said about noise figure relating to low noise transistors?

So, today, I'll try to take you through the noise considerations.

ERROR: All the above is in error by 7.5dB or a factor of 2.4, and that means the lowest output of the tape head is nearer 58uV than 24uV.

Reminder to myself: wake brain up first Wink



Edited by Graham Slee - 28 Nov 2022 at 11:05am
My blog is sponsored by Cadman Enterprises Limited
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Online
Points: 15025 		Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 28 Nov 2022 at 10:21am
Now, before we go there, it's best if you see the tape head response so you know the two sum to make flat.

I don't have a tape head or reel-reel to demo this, but by flipping the above image gives as near as damn it, the tape head output...

tape head output illustration

I have cropped the image at 50Hz to the left, and about 12kHz to the right, so it looks the right shape as you'd see in tape head manufacturers' charts.
My blog is sponsored by Cadman Enterprises Limited
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Online
Points: 15025 		Post Options Post Options   Thanks (1) Thanks(1)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 28 Nov 2022 at 10:34am
And here I found this image from Nortronics describing what the replay response should be (ignore the REC curve. The PB curve is what we want). And it looks like the difference between 50Hz and 1kHz is approximately what we have, within one or two dB. And we have to tighten that up a little, maybe.

Nortronics tape head response
My blog is sponsored by Cadman Enterprises Limited
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Online
Points: 15025 		Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 28 Nov 2022 at 11:56am
Now, the reason for this drawn out preamble, is that I think the noise performance isn't all that great, and as is usual, we look for trade-offs.

If the low frequencies are rolled off then the gain at low frequencies is less, and the low frequency noise is lower.

But what is acceptable?

Currently, and based on manufacturers data, we are a smidgen low at 100Hz, less than 0.5dB down at 100Hz.

We are led to believe that on a 6dB/octave slope, which this is, that 50Hz is 6dB larger than 100Hz (which it would be normally).

So, if the 100Hz to 1kHz level difference is 20dB - that's 20dB per decade which is 6dB per octave - then an octave higher must be 26dB.

However, 50Hz is an inflection point, and is 3dB higher, not 6dB.

What we also know about 3dB inflection points, is that at double the frequency (low pass), it hasn't become 0dB but minus 1dB, not nearly as neat as illustrative drawings show.

So the last image showing the curve passing through 100Hz at 20dB higher than 1kHz is untrue. It should be 19dB.

Therefore, as we cannot plot 50Hz accurately due to gain limitations of this amplification stage, then if 100Hz reaches +19dB, it can be argued that the 50Hz pole will be correct, but might be very slightly down against test tapes.

But my experience in phono EQ suggests that such a tiny amount will never be heard.

As such, we may be able to reduce the gain slightly to reduce the noise.


Edited by Graham Slee - 28 Nov 2022 at 1:28pm
My blog is sponsored by Cadman Enterprises Limited
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Online
Points: 15025 		Post Options Post Options   Thanks (1) Thanks(1)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 28 Nov 2022 at 12:13pm
So, by reducing the value of the 4.7 meg resistor in the tape EQ circuit to 3.3 meg, simulation suggests that at 50Hz, it will be 1.2dB down, and at 100Hz, it will be 0.6dB down.

How sensitive are we to level changes at low frequencies? I'll leave you to study the Fletcher and Munsen curves.

There will be those who will argue to the smallest fraction of a decibel, but this is about practicality - what can be achieved.

Therefore, I will try 3.3 meg-ohms and do the noise and distortion measurements and report back later.
My blog is sponsored by Cadman Enterprises Limited
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Online
Points: 15025 		Post Options Post Options   Thanks (1) Thanks(1)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 28 Nov 2022 at 3:29pm
What noise level is regarded normal for reel to reel tape recorders?

I had a quick look at a few randomly selected machines on reel-reel.com, and they vary from 45dB - 55dB. There was one at 65dB but it said weighted, which means the lows and highs are filtered out similar to the Fletcher and Munson curves, hence the better figure.

So, with this circuit, I decided to set the sensitivity to 1mV at 1kHz which gave about 405mV output, and then simply switched the signal generator off.

The average reading was around 1.5mV, and that's noise.

405/1.5 = 270. The log of 270 is 2.43, and multiplied by 20 to obtain dB, is 48.6dB.

Now, that isn't with any filters and we usually measure 20Hz - 20kHz, so I'd expect it to be better than 48.6dB.

So I set up for the F and M curve weighting, which is roughly A.wtd and added a 100Hz high pass filter and got 69dB ref 1kHz.

tape head preamp noise A.wtd

Then I set it to no high pass filter and no weighting, and to emulate a real tape signal at 50Hz, reduced the signal by 22dB, and got nearly 51dB.

tape head preamp noise -22dB input at 50Hz

Then I set the generator level with the average 1mV head output, so that the boosted low frequencies were averaged with the higher un-boosted frequencies, which means around -7dB, so that's about 400uV, and got the THD sweep plot below, showing average distortion around 0.1%.

tape head preamp THD sweep

And before I forget, I did a frequency sweep which confirms the simulation, that at 50Hz it is 1.2dB low, which means it is 0.6dB low at 100Hz.

tape head preamp EQ frequency response sweep

And also measured the 1kHz to 10kHz frequency difference, which I need to go away and look up to see what it should be, and I'll come back to that later.
My blog is sponsored by Cadman Enterprises Limited
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Online
Points: 15025 		Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 28 Nov 2022 at 11:29pm
As a matter of interest, I'm basing the gain of the tape stage on head information from here:

Nortronics Tape Heads: 4 track; 1/4&quot; tape

If you think I'm in error please say.
My blog is sponsored by Cadman Enterprises Limited
Back to Top
 Post Reply Post Reply Page  <1 255256257258259 281>
  Share Topic   

Forum Jump Forum Permissions View Drop Down

Forum Software by Web Wiz Forums® version 12.01
Copyright ©2001-2018 Web Wiz Ltd.

This page was generated in 0.094 seconds.