User avatar
EmAtChapterV
Posts: 244
Joined: Sun Jan 06, 2013 6:49 pm
Location: Vancouver, BC

Tank circuit to attenuate cutterhead resonance?

Post: # 22714Unread post EmAtChapterV
Tue Jan 22, 2013 11:32 am

An electrical theory noob, I'm sitting here pondering what would be involved in building a Tank Circuit (coil and capacitor in parallel) to flatten out the 4.5 kHz resonance peak in my Presto 1D. Power-wise it seems wasteful at first glance, but since neither the coil nor the capacitor are really resistors, it's just the amp "sees" them as such together at a certain frequency. So is power being wasted, or was it never there to begin with? It seems like the latter, but then again I don't want to eliminate everything at the resonance, just attenuate it 24 dB or so, which would involve putting a resistor in parallel with the coil and cap. And that resistor will be doing work converting power to heat because IIRC that's what resistors do, and I need to figure out what the value of this resistor ought to be vis a vis the reactance of both the head and the tank circuit, and aaargh... :?

I feel like such a slow learner. I'm getting the idea that something involving a negative feedback op amp between the source material and the amplifier is a far more efficient way to go, but it seems beyond my skill level at the moment.

At least I figured out what was causing the mysterious clipping. For some strange reason my amplifier goes from flat at 16 kHz to nearly 6 dB up at 20 kHz, which coincidentally is nearly perfectly matched by losses in the impedance-matching transformer. It does this into a dummy load as well, so it's not the transformer's fault. I suspect it may need servicing, and also that I'm asking it to do things it was never designed to do.

User avatar
dubcutter89
Posts: 359
Joined: Thu Oct 19, 2006 6:30 am
Location: between the grooves..

Re: Tank circuit to attenuate cutterhead resonance?

Post: # 22716Unread post dubcutter89
Tue Jan 22, 2013 1:18 pm

...flatten out the 4.5 kHz resonance peak in my Presto 1D ... attenuate it 24 dB or so...
24dB ??? - if the peak is really that heavy then you should check your heads internal damping system...

anyway, what you are looking for is a "notch-filter".
an easy way would be a series circuit of RLC. other ways are aktive (opamp) notch-f., or twin-T....
i suggest to use a RLC series notch-filter in the line signal before the amp.
when you use a potentiometer instead of a fixed resistor you can dial in attenuation from zero to ...whatever...

Image

now use a potentiometer (maybe 10k or 100k) instead of R1 R2 and you have variable filter
the triangle with zero connects to signal ground.

f = 2 Pi / SQRT (L*C) ...all formulas are anywhere online...

hope this helps

lukas
Wanted: ANYTHING ORTOFON related to cutting...thx

User avatar
EmAtChapterV
Posts: 244
Joined: Sun Jan 06, 2013 6:49 pm
Location: Vancouver, BC

Re: Tank circuit to attenuate cutterhead resonance?

Post: # 22718Unread post EmAtChapterV
Tue Jan 22, 2013 2:15 pm

This is 24 dB over the entire sonic spectrum - probably closer to 18 dB if you take into account FFT analysis weirdness and transformer losses in the bass frequencies. (I'm waiting on a better transformer to arrive in the mail, a Hammond 119DA.) I'm trying to flatten it with subtractive EQ only, with an eye towards using passive components when it comes time to build the actual physical version. But maybe this isn't practical.
You do not have the required permissions to view the files attached to this post.

User avatar
markrob
Posts: 1624
Joined: Wed Nov 28, 2007 1:14 am
Location: Philadelphia Area

Re: Tank circuit to attenuate cutterhead resonance?

Post: # 22720Unread post markrob
Tue Jan 22, 2013 3:29 pm

Hi,

You should do any EQ at line level audio prior to the power amp not at the head. Much eaiser to build (passive or actively or do digitally). How are you measuring this? Are you actually cutting a lacquer and and this is the playback of a cut? Did you cut a white or pink noise signal or is this a frequency sweep. Was RIAA encoding used prior to cutting? RIAA on playback? The RC network detailed in the 1D manual gives to the 500hz turnover specified in the RIAA spec. So you will see a drop off in frequency below 500 Hz.

Mark

User avatar
EmAtChapterV
Posts: 244
Joined: Sun Jan 06, 2013 6:49 pm
Location: Vancouver, BC

Re: Tank circuit to attenuate cutterhead resonance?

Post: # 22725Unread post EmAtChapterV
Tue Jan 22, 2013 8:21 pm

markrob wrote:Hi,

You should do any EQ at line level audio prior to the power amp not at the head. Much eaiser to build (passive or actively or do digitally). How are you measuring this? Are you actually cutting a lacquer and and this is the playback of a cut? Did you cut a white or pink noise signal or is this a frequency sweep. Was RIAA encoding used prior to cutting? RIAA on playback? The RC network detailed in the 1D manual gives to the 500hz turnover specified in the RIAA spec. So you will see a drop off in frequency below 500 Hz.

Mark

This is about a minute of white noise cut to lacquer, flat - no RIAA curve on recording or playback - at 33 1/3 rpm at about 9 inches diameter, with hot stylus set to 0.4 amps. It's played back with an Ortofon OMB 10, corrected to perfectly flat response in both channels give or take less than 1 dB, using the white noise track on the Hi-Fi Sound HFS-75 test LP (cut by an Ortofon DSS731) for calibration.

And you're both absolutely correct about line-level EQ. I don't know why I'd gotten it into my head that anything line-level had to be active, not passive. Sometimes the simplest answer is right under my nose and I can't see it.

User avatar
markrob
Posts: 1624
Joined: Wed Nov 28, 2007 1:14 am
Location: Philadelphia Area

Re: Tank circuit to attenuate cutterhead resonance?

Post: # 22726Unread post markrob
Tue Jan 22, 2013 9:14 pm

Hi,

The system resonance seem too low for this head. Should be up in the 8Khz range. Also seems very under damped. Have you checked to see if the rubber damping material is old and dried out? Also check to make sure the small discs that couple the damping rubber is properly clamped to the armature. The small brass screws may be loose or stripped. Check to make sure the clamping mechanism has not broken free from the armature. The bottom disc is screwed and soldered to the aramture. If this is loose then it won't be able to damp the head. Are you using the long shank stylus? If so, try a short shank. This is stiffer and might improve the high freq performance (low output would be the drawback here). One other area of concern are the two balancing springs. If they have lost pressure, that would reduce the stiffness and lower the system resosnance. To test velocity response of the head, you may be better off running a light band test as shown in the Presto manual. This eliminates any calibration issues on playback. Fro beast results run this at 78 rpm at a low tpi (maybe 120).

Mark

User avatar
EmAtChapterV
Posts: 244
Joined: Sun Jan 06, 2013 6:49 pm
Location: Vancouver, BC

Re: Tank circuit to attenuate cutterhead resonance?

Post: # 22727Unread post EmAtChapterV
Tue Jan 22, 2013 11:51 pm

Well, crud. There's the problem: the bottom/ inner disc on the armature wasn't just loose, it had corroded enough that it had a crack in it, running through the center of two of the tiny screws. It fell apart when I unscrewed them; now I have 1 2/3 damping discs and a non-operational cutterhead. I'd ask if anyone has spares, but I'm thinking a proper full service by someone who knows what they're doing is likely called for. The coils are good but everything else is probably cattywampus.

I'm really disappointed in myself right now. I'm terrified of breaking things even further.

User avatar
EmAtChapterV
Posts: 244
Joined: Sun Jan 06, 2013 6:49 pm
Location: Vancouver, BC

Re: Tank circuit to attenuate cutterhead resonance?

Post: # 22730Unread post EmAtChapterV
Wed Jan 23, 2013 1:05 am

Pictures of the problem...
You do not have the required permissions to view the files attached to this post.

User avatar
markrob
Posts: 1624
Joined: Wed Nov 28, 2007 1:14 am
Location: Philadelphia Area

Re: Tank circuit to attenuate cutterhead resonance?

Post: # 22745Unread post markrob
Wed Jan 23, 2013 9:37 am

Hi,

At least you know why you were not getting much damping action. Looking at the picture, I think you can salvage this. Just screw the bottom disc back on the head and secure it with the red Loctite 271 (this is more pernament than the blue stuff). Make sure you clean up the threads on both surfaces before locking into place. When you re-install the bottom disk align it so that the the old holes are not in line with the holes in the rubber. Using the rubber as a drill guide, drill and tap 3 or 4 (if you are able) new holes into the bottom disc. Then re-assemble and you are good to go. You can probably get by with using only 3 holes. I used 0-80 brass round head screws. You can get them along with drill and taps here:

http://www.micromark.com/round-head-machine-screws-0-80-x-3and8-inch-pkg-of-10,7298.html

http://www.micromark.com/tap-0-80nf,6514.html

If you want a complete repair, make a new disk then tap and drill as stated above.

Your other option would be to remove the damping altogether and just use EQ to flatten the head response.

Befrore you deal with this, you'll have to figure out why the resosnsce is so low. I believe this is either due to tired or under tensioned armature springs. These can be adjusted after unlocking the springs using two set screws on the back of the head. You adjust the springs using the set screws on the side of the head. Be careful when making the adjustment as its easy to snap the set screws if you overdo things. Try a 1/4 turn at a time and see if the resonance is increased. When you make this adjustment, you'll also need to re-center the aramture in the gap. If you tighten one of the screws, you need to do the same to the other screw to force the amature back to center. Maybe its already out of adjustment. If you have the manual, you can read more about this.

Mark

User avatar
Radardoug
Posts: 58
Joined: Fri Sep 23, 2011 4:22 am
Location: New Zealand
Contact:

Re: Tank circuit to attenuate cutterhead resonance?

Post: # 22760Unread post Radardoug
Wed Jan 23, 2013 2:49 pm

EQ'ing to get rid of a resonance problem is never a good idea. The resonance is always there and will affect the signal. Much better to correct the resonance.

User avatar
markrob
Posts: 1624
Joined: Wed Nov 28, 2007 1:14 am
Location: Philadelphia Area

Re: Tank circuit to attenuate cutterhead resonance?

Post: # 22768Unread post markrob
Wed Jan 23, 2013 9:06 pm

Hi,

I see where you are coming from. However its not really a "problem" with the head but is unavoidable due to the physics. The issue with is the difficulty of controlling the physical properties of the damping material so that it acts as a true mechanical equivalent of a resistor in an electrical circuit. Its also prone to change vs. temperature and over time. On the electrical side, you have the problem of matching a correction filter to the resonance peak, which can also vary over time, temperature, and mechanical load. I think in either case, if you get it correct the end result is flat response. Consider that some of the moving coil designs do not employ any mechanical damping, but use motional feedback to flatten the response. This could be viewed as electrical damping.

Mark

User avatar
EmAtChapterV
Posts: 244
Joined: Sun Jan 06, 2013 6:49 pm
Location: Vancouver, BC

Re: Tank circuit to attenuate cutterhead resonance?

Post: # 22772Unread post EmAtChapterV
Wed Jan 23, 2013 9:46 pm

The springs were very loose, I've tightened them up a bit and made sure to re-center the armature. I've put everything back together, and am waiting on the threadlock to set and cure. I'll run some cutting tests tomorrow and post results.

User avatar
markrob
Posts: 1624
Joined: Wed Nov 28, 2007 1:14 am
Location: Philadelphia Area

Re: Tank circuit to attenuate cutterhead resonance?

Post: # 22779Unread post markrob
Thu Jan 24, 2013 8:06 am

Hi,

That sounds promising.

One quick way to see if you have made an impovment is to hook the head to a scope (or the input of a computer sound card). Install a 1/16" short aluminum rod or wire to take the place of the cutting stylus. Tap or tweak the end of the rod and to induce an impulse. Look at the frequency of the damped sine wave generated by the head on the scope and you'll be able to get a good estimate of the system resosnance and how well damped it is. That will save quite a bit of time cutting and wasting of blanks.

Mark

User avatar
EmAtChapterV
Posts: 244
Joined: Sun Jan 06, 2013 6:49 pm
Location: Vancouver, BC

Re: Tank circuit to attenuate cutterhead resonance?

Post: # 22796Unread post EmAtChapterV
Thu Jan 24, 2013 7:31 pm

Well, that was unexpected: after the improvements, the resonant frequency hump is now lower and steeper than before. :? Sensitivity has increased about 2 dB at 1 kHz, but everything else has gotten worse.

Here are the graphs, plus two sets of light band tests at 78 rpm - eleven spot frequencies of 1, 1.25, 1.6, 2, 2.5, 3.15, 4, 5, 6.3, 7.9 and 10 hz, about 3 seconds each. The outer band was with the stylus heat on, the inner was with it off.

Time to take the head apart again, readjust and try again... Though I'm beginning to wonder if age may have caught up to the springs. I'll try connecting the head to computer's sound input and tapping the armature this time.
You do not have the required permissions to view the files attached to this post.

User avatar
EmAtChapterV
Posts: 244
Joined: Sun Jan 06, 2013 6:49 pm
Location: Vancouver, BC

Re: Tank circuit to attenuate cutterhead resonance?

Post: # 22804Unread post EmAtChapterV
Fri Jan 25, 2013 12:10 am

Disregard the previous graph - for some reason I had a high-pass filter switched in while cutting the white noise tests. :oops: Here's the correct one. Most behaviour above 1 kHz, including the lowered resonance, remains the same as the previous post, though. I beat-cancel checked the lathe's turntable speed against a quartz-locked direct-drive Technics deck; it's spot-on give or take less than 0.1%, the same as ever.
You do not have the required permissions to view the files attached to this post.

User avatar
markrob
Posts: 1624
Joined: Wed Nov 28, 2007 1:14 am
Location: Philadelphia Area

Re: Tank circuit to attenuate cutterhead resonance?

Post: # 22811Unread post markrob
Fri Jan 25, 2013 9:55 am

Hi,

That's a drag. It seems there is a issue with the springs as I doubt there is any extra mass coupled to the head. What happens if you test the head with the damping rubber removed? Have you looked at the spirngs? Is it possible somebody replaced them at some point? They should have an "omega" shape to them. Also, since they were to be loose, how much did you tighten them? Did you do anything to the center adjustment that is used to force the knife edge of the aramture bearing in solid contact with the V groove? These heads are pretty well designed and easy to work on. So I think you should be able to get it working well. If you have to make new springs, its not too bad. I did this on one of my heads. IIRC I used .039" piano wire. The trick was getting it bent into the shape and length needed. Hopefully, you'll be able to get it solved without resorting to that. I'm not sure if Gib still does these and if he knows how bring them back into spec.

Mark

User avatar
Radardoug
Posts: 58
Joined: Fri Sep 23, 2011 4:22 am
Location: New Zealand
Contact:

Re: Tank circuit to attenuate cutterhead resonance?

Post: # 22820Unread post Radardoug
Fri Jan 25, 2013 3:57 pm

The problem is a resonant circuit has a time component. So you can use an antifilter to flatten the steady state response, but this does not mean that when excited by music, the response is flat. The resonance will induce time related problems. If you can reduce the resonance then you will improve the end result which is after all a music signal. The mechanical damping does reduce the time element.
And yes, physics is at work, and you can't beat physics!

User avatar
markrob
Posts: 1624
Joined: Wed Nov 28, 2007 1:14 am
Location: Philadelphia Area

Re: Tank circuit to attenuate cutterhead resonance?

Post: # 22821Unread post markrob
Fri Jan 25, 2013 5:29 pm

Radardoug wrote:The problem is a resonant circuit has a time component. So you can use an antifilter to flatten the steady state response, but this does not mean that when excited by music, the response is flat. The resonance will induce time related problems. If you can reduce the resonance then you will improve the end result which is after all a music signal. The mechanical damping does reduce the time element.
And yes, physics is at work, and you can't beat physics!
Hi,

I think I'm correct in stating that if the system representing the head can be classified as minimum phase, then applying the inverse will result in a flatened response and remove any time related errors due to group delay.

That is H(s) = (Z(s) / P(s)) x (P(s)/Z(s)) = 1 Where Z(s) and P(s) are the poles and zeros of the system representing the head.

The question is can you consider the head to be a minimum phase system?

User avatar
EmAtChapterV
Posts: 244
Joined: Sun Jan 06, 2013 6:49 pm
Location: Vancouver, BC

Re: Tank circuit to attenuate cutterhead resonance?

Post: # 22830Unread post EmAtChapterV
Fri Jan 25, 2013 7:29 pm

Sort of an aside, I accomplished (?) something this afternoon I haven't before in a decade and a half of cutting. I was setting up to do a couple of music quality tests on the head before I pulled everything apart again...

"In the stillness of remembering, what you had, And what you lost, And what you had, And what you lost. Thun" (SNAP) (fizzle). :shock:

Apparently 63 milliamp fuses don't like Stevie Nicks. Who knew? :roll:

Post Reply