V1.0 of Bryan's 3D printed head with feedback

Anything goes! Inventors! Artists! Cutting edge solutions to old problems. But also non-commercial usage of record cutting. Cost- effective, cost-ineffective, nutso, brilliant, terribly fabulous and sometimes fabulously terrible ideas.

Moderators: piaptk, tragwag, Steve E., Aussie0zborn

User avatar
markrob
Posts: 1623
Joined: Wed Nov 28, 2007 1:14 am
Location: Philadelphia Area

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 31635Unread post markrob
Wed Oct 01, 2014 4:50 am

Hi,

1. Run the same set of tests at 20Khz.

2. Run the same set of tests mechanically coupled (1 and 20 Khz).

3. Make sure the coupled output is at least 5-10db higher than the uncoupled. If so, step 4, else more shielding is needed.

4. Drive with pink noise (if using octave based spectrum analyzer) and view your open loop response. See if you get correct response. View system resonance.

Mark

User avatar
EpicenterBryan
Posts: 738
Joined: Sun Aug 10, 2014 9:01 pm
Location: Eugene, OR USA

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 31699Unread post EpicenterBryan
Thu Oct 02, 2014 11:54 pm

A mini update:

I ordered some mu-metal like material today (Ultraperm 80) to do some shielding at some point and as I was thinking about the best places to apply it and I had a DUH! moment. I realized I still had 4 metal rods holding the drivers and the rest of the stuff on each driver / sensor pair together. I had planned to replace them with nylon and forgot. So along side the driver and feedback transducer were not one, but 4 possible magnetic linkages. To make matters worse, I started to collect some data and realized my hot shot old school true RMS meter (that hasn't been through the calibration lab for decades) has a zero offset - not too dramatic, but I don't trust the previous data other than for order of magnitude like readings.

It's not a complete loss... I know now that the push rod needs to be non-magnetic. So I will regroup and take new data and with the recommendations of Mark, I will gather data for a few extra frequencies at the same time.

I'm also going to look into other things like changes to the threaded disc design to reduce some moving mass since I need that disc on not only the driver but also the feedback unit, and I'm going to use super glue to attach it rather than epoxy which will also help. Now that I've taken the cone and surround off a driver (but left the spider in place) for testing the magnet coupling I realize that for the feedback coil I don't need that anyway and the disc connecting to the push rod doesn't need to be as robust as the driver. Also, the driver disc only really needs to be as thick as it is where it's threaded for the push rod, and not the rest of it. Little changes like this add up.

Tonight, after a long discussion with Tom from Taulman 3D about the nylon "Bridge" product, I'm drying the filament in a food dehydrator which I modified today. It will take 24 hours to dry out. Hopefully I can print a new head body tomorrow with some additional speed and retraction changes as he recommends for my application. But I may not be successful until some super small drill bits arrive and I can custom bore some different nozzle sizes that are not normally made for my extruder.
IMG_2942[1].JPG
You do not have the required permissions to view the files attached to this post.

User avatar
EpicenterBryan
Posts: 738
Joined: Sun Aug 10, 2014 9:01 pm
Location: Eugene, OR USA

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 31739Unread post EpicenterBryan
Tue Oct 07, 2014 12:52 am

I spent a very frustrating weekend messing with the Nylon Bridge product. I had about an hour long chat with Tom from Taulman 3d about the product and we went over all my settings and hardware last Thursday. As a result I dried the product for 36 hours over Thursday and Friday. Over the weekend I tried everything on the list and the results are shown below. The only thing I could not test was to use a larger diameter nozzle. 4mm is the largest they make for my print head. So I have drill bits on order and spare undersized nozzles on order so I can bore them out to custom sizes and will try not only 0.5 but also 0.6, 0.8 and 1.0mm. (standard is 0.4mm as was used on the YouTube video where it worked so great).

In all fairness, I have to say that Tom was very responsive and interested in the problems once we got past the useless E-mail I got on my first contact, so big plus there.

The bottom photo shows the part I tried to print in Nylon Bridge, and the same successful part I printed Sunday night in my normal ABS plastic right above it. Night and day.
Comp2.jpg
So I gave up for the weekend and on Sunday night I printed a new version of the bracket in ABS plastic that would fit my presto 8D, but also my K11 (as I thought at the time). In ABS it was a 5 1/2 hour print. In the Nylon Bridge product it would be about a 9 hour print.
Comp1.jpg
No I have not given up on this stereo with feedback project.... I think it has merits but needs more testing. I'm waiting for some Mu-Metal to arrive for the stereo version WITH feedback, and for my Caruso boards as well. But for now I have decided to lock down a stereo version with NO feedback, with the body printed in ABS plastic as a baseline so I can see how this works. I still think a Nylon body would work best since it would have slight properties like the "presto flex" but I can't get the stuff to print yet.

In this photo you see the stereo (no feedback) head mounted on my K11, and the Stereo With feeback in the background. Also the stock Presto head.
IMG_2947.JPG
Oh, but the mounting hole locations on the 5C head (stock for a K11) is not the same as a 1C / 1D. So all the sudden I need to make an adapter plate (which isn't needed on my 8D lathe) to mount this head to a suitcase Presto. ARRRRRGGGGGHHHHH!

IMG_2948.JPG

More to come.... Just not enough hours in the day.

Bryan
You do not have the required permissions to view the files attached to this post.

User avatar
opcode66
Posts: 2700
Joined: Sun Nov 08, 2009 10:56 pm
Contact:

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 31740Unread post opcode66
Tue Oct 07, 2014 2:24 am

I had the same sort of issues. I spent so much time on my overall design and on the design of the transducers that I simply overlooked certain aspects regarding mounting and passage over the disc. Once I mounted it, however, these considerations were painfully obvious. But, sometimes, you can't see the forest for the trees...

As you know, it really sucks to have to make changes to the cutterhead body itself. Tweaking parts, no problem. But, as it stands right now, my cutterhead body takes 6 hours to print. That is a lot of time and plastic. I've therefore tried to be very sure I had all my changes made appropriately on the body before printing. Pulling the trigger on those prints is rather nerve wracking. Currently, I've resolved all but one issue with respect to my mounting. To fix the last one, I will have to make a change to the body that impact the design of four other major parts. Not looking forward to all of that.

I know it is frustrating to run into issue like this. But, it is a sense of accomplishment to actually be this far. So, let that sink in and simmer for a while. You are having these issues because you've come this far. And, that is a good thing overall. Keep up the good work. Let's keep those turntables turning.
Cutting, Inventing & Innovating
Groove Graphics, VMS Halfnuts, MIDI Automation, Professional Stereo Feedback Cutterheads, and Pesto 1-D Cutterhead Clones
Cutterhead Repair: Recoiling, Cleaning, Cloning of Screws, Dampers & More
http://mantra.audio

User avatar
EpicenterBryan
Posts: 738
Joined: Sun Aug 10, 2014 9:01 pm
Location: Eugene, OR USA

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 31755Unread post EpicenterBryan
Wed Oct 08, 2014 11:22 pm

IMG_2966.JPG
Calling all trolls:

I made the adapter made to mount the No Feedback version on my K11 for the first round of testing. I'm running out of time tonight but I have files ready to do a test cut. Before I do, I want to list what I'm planning to cut and see if any of you guys think I am missing something. I have prepared these files with "Audacity" which has surprisingly great features, and I'll be using it for FFT of the test cuts.

Here is what I have prepared and plan to cut. All files are 30 seconds long. The sweeps and Pink Noise files have the RIAA curve applied. There will also be plenty of silent grooves as well.

1). 1khz mono normalized at 0db.
2). 1khz mono at -20db.
3). Sweeps:
  • a. MONO 20-20khz with RIAA applied, normalized to 0db
    b. LEFT ONLY 20-20khz with RIAA applied, normalized to 0db
    c. RIGHT ONLY 20-20khz with RIAA applied, normalized to 0db
4). Pink Noise
  • a. MONO Pink Noise RIAA applied, normalized to 0db
    b. LEFT ONLY Pink Noise RIAA applied, normalized to 0db
    c. RIGHT ONLY Pink Noise RIAA applied, normalized to 0db
    d. STEREO Pink Noise RIAA applied, normalized to 0db
I'm not sure how useful the sweeps are, since I plan to use Pink Noise with my spectrum analyzer and also with the functions built into Audacity but I made the files and can cut them if anyone thinks they may be of value.

I also realize this swing arm K11 is not the ideal test bed since it is not linear tracking, and I have made things worse by having the stylus further forward in this stereo head (compared to the mono head), and adding an extra 8mm adapter block doesn't help either, but it's better than tearing my 8D apart to mount it or building a custom overhead (for now).

Thoughts, comments and such?

Bryan
You do not have the required permissions to view the files attached to this post.

User avatar
markrob
Posts: 1623
Joined: Wed Nov 28, 2007 1:14 am
Location: Philadelphia Area

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 31761Unread post markrob
Thu Oct 09, 2014 7:22 am

Hi,

Looking froward to seeing your results. If you do RIAA sweeps, watch the levels. One thing to consider is building up a flat phono preamp (no RIAA) for testing. This lets you measure the velocity response of the your head directly if you use a typical magnetic pickup for playback. I built one with a gain of 100 and the proper 47K ohm input impedance (also be sure to add the correct total capacitance specified by your pickup maker).

The geometry of your head as mounted in the photo looks really bad. Perhaps its the exaggerated due to the perspective in the photo. If its that far off, you will have quite a bit of tracking error on playback. It may be ok to get a first look, but you'll need to correct this.

Mark

User avatar
rsimms3
Posts: 427
Joined: Thu Mar 14, 2013 10:01 am

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 31762Unread post rsimms3
Thu Oct 09, 2014 9:04 am

I was wondering about the geometry as well. The space/adapter appears to be close to the same depth as the rear of a 5c head but the use of the drivers puts the stylus father out from the mount than with a mono head.

Image

User avatar
EpicenterBryan
Posts: 738
Joined: Sun Aug 10, 2014 9:01 pm
Location: Eugene, OR USA

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 31783Unread post EpicenterBryan
Sun Oct 12, 2014 10:41 pm

markrob wrote:The geometry of your head as mounted in the photo looks really bad.
Yes, you guys were right. I started looking into all the turn table tracking error stuff (which applies to this swing arm) and decided it could be partially corrected for with an angled mounting block but decided to mount the head to my 8d instead so I could get on with it. I really wanted to see what the response of the head was like and if it was possible.

I spent a glorious day playing, testing and learning - just me and the gear. I'm going to skip a bunch of the details and get right to the highlights.

First off, the plastic head weighs 5.4oz. A Presto 1D head weighs is 9.9oz, and a Presto 5C head is 8.1 oz. I decided to leave the spring set where it was for the 1D head and add weight to the plastic head to make it the same as a 1D until a scale I have on order arrives and I can measure the cutting down force and make proper adjustments. Shown here is the plastic head with 3 metal plates added on the front to make the weight the same as an 1D head.
IMG_2969.JPG
After chasing my tail trying to debug some really poor response initially, I decided to eliminate as much gear as possible. So I ended up using an Alesis IO2 for direct output from the PC to the Phase Linear Amp. I also used it to measure the playback through the RIAA phono preamp. Ultimately, it turned out the files I prepared to burn to a CD to provide the test files for some reason did not have the RIAA pre-compensation curves applied so I made them again and triple checked them. These updated files were used to drive the amp directly from the Alesis. SOOOOO much better!!! I'm sure what I had set up initially would have worked fine if I was outputting the correct files, but this is a better solution anyway with less gear coloring things anyway.

The next issue was that I had the cutting angle set totally wrong. Rather than a setback angle of 5 degrees, it was actually set forward which explained a bunch of issues. But it also brought up an interference issue with the adapter which left very little clearance on the back side between the head housing and the disc surface when the correct back angle is used. It just barely cleared. It's an easy fix with a mounting adapter design change.
IMG_2970.JPG
Another "gotcha" on my project also got Todd on his: I discovered the body will need a modification so it can travel the full distance to the center especially when the vacuum disc hold down adapter is in place. Another duuuuhhhh! I'll have to look into this one...
IMG_2976.JPG
Due to the number of upload files per post, I'll continue this in another one...

Bryan
You do not have the required permissions to view the files attached to this post.

User avatar
opcode66
Posts: 2700
Joined: Sun Nov 08, 2009 10:56 pm
Contact:

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 31784Unread post opcode66
Sun Oct 12, 2014 10:48 pm

FYI, on a Neumann the cutterhead leans forward. That is actually what I'm intentionally going for. It is totally possible to cut that way.

Keep up the good work.
Cutting, Inventing & Innovating
Groove Graphics, VMS Halfnuts, MIDI Automation, Professional Stereo Feedback Cutterheads, and Pesto 1-D Cutterhead Clones
Cutterhead Repair: Recoiling, Cleaning, Cloning of Screws, Dampers & More
http://mantra.audio

User avatar
EpicenterBryan
Posts: 738
Joined: Sun Aug 10, 2014 9:01 pm
Location: Eugene, OR USA

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 31785Unread post EpicenterBryan
Mon Oct 13, 2014 12:54 am

OK, here is what you really wanted to see....

This is the response of a single frequency test tone of 1khz, referenced at digital 0. This is the one and only frequency input and the resulting output. Before you dive into this too much, let me make sure you understand that I have not totally resolved my speed issue on my 8D with my variable frequency drive. I'm still waiting for a more accurate strobe and a multi turn pot so I can really dial in the frequency. So for now, just assume the signal near 1khz in the chart is correct (even though the actual test file is exactly 1.000khz).
1khz response.JPG
On the right side of 1khz I expect to see odd and even harmonics all decreasing - the 3rd harmonic is much higher than I would expect. But what is on the left side of 1K is not in the source file and is obviously related to resonances, vibration from the turntable / lathe coupled to the head, wind noise from the vacuum system or pipe. I'm not clear what to think just yet. But you will see a similar bump in that range in the carts to follow.

Before we go too far into this, unanticipated resonances in materials can always mess things up. I mentioned a few weeks ago that I could hear this head ring when I scratched the side with my fingernail. I just didn't know how bad it would be when actual stimulus was applied. In the studio world you guys know that parallel walls and right angles are bad and can create standing waves and goofy room responses. Now that I see how this head responds, I'm going to think of this it as a room that needs mods so things don't resonate.

Below is another interesting clue. This is the 1Khz output zoomed out. You will notice a low frequency modulation in the actual output. My gut also says that the head may have a resonance that is very close to the 1khz test input as is implied by this visual beat frequency. Yes, you can hear the 1khz playback increasing and decreasing in frequency. The time scale is in seconds.
beat frequency.JPG
Since the head is 3D printed, there is nothing that says I can't apply crazy patterns like diamond plate in interior or exterior surfaces that do not need to be flat, and there is nothing that says I have to keep corners at 90 degrees except where needed to mount things.

The next thing I want to try is Nylon Bridge for the main body when I have the ability to make larger nozzles for my printer, and at the same time I'm going to try to break up the inner and outer surface (but especially the corners) to eliminate any possibility of standing waves being developed.

Here is the thing you really want to see. This is the 20kz to 20Khz response.
head response.JPG
It looked pretty bad until I thought about the resonances that my be different with a different print material, or by intentionally messing with the surface and the corners of the print. I have a few ideas what to do about 7Khz and up, but will need to play with that...

It does look like a parametric EQ could fix the 3 big issues, but I would rather fix the 3D printed housing that seams to be the root of the issue.

Bryan
You do not have the required permissions to view the files attached to this post.

User avatar
EpicenterBryan
Posts: 738
Joined: Sun Aug 10, 2014 9:01 pm
Location: Eugene, OR USA

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 31786Unread post EpicenterBryan
Mon Oct 13, 2014 1:15 am

I should also note that I made changes to the discs that glue to the drivers. This reduced the moving mass. But I still have a steel 3mm screw holding the stylus since my nylon screws have not yet arrived. Here is an updated internal view:
IMG_2979.JPG
Also, I took Todd's advice and used super glue to attach the new disc to the driver. And all the components on this version had Lock Tite "Blue" applied as I assembled things. I have no clue if I can ever take it apart.
You do not have the required permissions to view the files attached to this post.

User avatar
EpicenterBryan
Posts: 738
Joined: Sun Aug 10, 2014 9:01 pm
Location: Eugene, OR USA

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 31787Unread post EpicenterBryan
Mon Oct 13, 2014 1:36 am

I forgot to attach the noise input file...
This has the RIAA curve applied to the noise source. Notice the high frequency is about 20db hotter than 1K, and the low frequency is about 20 db lower than 1Khz just as it should be.
noise input.JPG
Also attached is the full frequency response in tabular form for those who want to mess with an EQ curve:
spectrum 10-12-14.txt
Bryan
You do not have the required permissions to view the files attached to this post.

User avatar
markrob
Posts: 1623
Joined: Wed Nov 28, 2007 1:14 am
Location: Philadelphia Area

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 31794Unread post markrob
Mon Oct 13, 2014 9:20 am

Hi,

Nice first results.

Here is what I see.

The 1Khz spectrum shows quite a bit of distortion (about 10%). Have you established what velocity level this represents? If you have a test record (e.g. CBS STR series), you can compare against that. The other way is to read the output of your pickup directly in millivolts AC. Most pickups specify the output at 1Khz 5 cm/sec. What is the relative playback level in db compared to this reference? Perhaps you cut at too high a level. If not, you need to find out where the source of non-linearity is (e.g. lost motion, sticktion, mounting geometry, etc.). You would like to see harmonics 30-40db down from peak (about 1% level). Also, make sure that you were not near clipping someplace in the signal path (either cutting or on playback). If you do get a test record, you can check you playback system to make sure it is not the problem.

On the frequency response front, the results at 1Khz are what you want to see (+6db /oct below resonance and -6db above it). On your sample there is some strange stuff happening at the low frequencies that I can't understand. You might want to cut a silent groove and get an idea of the system noise floor and spectrum. Do you have some AC mains signal (either electrical or mechanical) present that is skewing your results? There may also be some rumble present as well. Also make sure your head mounting has proper damping and is not fluttering. If you see similar junk when cutting a silent groove, you can work on reducing it so you can get good measurements. It will also help you to determine if your high frequency noise floor obscures your cutting results. I see that you have a fairly large secondary resonance at about 7Khz. This needs to bee addressed or it will kill your high frequency response. From my experience, this occurs in the coupling from the speaker to the stylus holder. In a perfect world, you would like the speaker to be in lock step with the speaker. However, in the real world, you have another spring mass system that gets in the way. Above the resonance, you decouple the stylus from the speaker, killing the high frequency performance. You may be able to kill the effect by adding some damping to the coupling members. Otherwise, a better design might be needed.

All in all, you seem to be on the way to a good result. Keep at it and I'm sure you'll get there.

Mark

User avatar
EpicenterBryan
Posts: 738
Joined: Sun Aug 10, 2014 9:01 pm
Location: Eugene, OR USA

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 31859Unread post EpicenterBryan
Sun Oct 19, 2014 1:10 am

markrob wrote:All in all, you seem to be on the way to a good result.
Mark, like usual you have really great technical comments and I hope to address them after this next round of tests. I have several game changer mods to apply and I need to take new data I think.

I thought it would be easier to explain and show progress in an unlisted YouTube video just for you guys.
A few things to note; in the tap test demo the head housings were placed on closed cell foam (a material called Poron) which was adhered to the concrete so you could hear the resonances. Also, the changes I made are about all that can be done (except for coating the housing as I mention) since I applied lock-tite as I assembled things. Depending on how things go in testing tomorrow, I may try "the dip" and acquire more data, but I think this will be the end for the ABS version. The next version will use nylon if I can ever get the stuff to print right. It should be an interesting day tomorrow.


Bryan

User avatar
emidisc
Posts: 220
Joined: Mon Jun 07, 2010 7:57 pm
Location: lancashire

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 31862Unread post emidisc
Sun Oct 19, 2014 10:50 am

Looking Good Bryan thanks for posting this, I have now recieved a few of these same drivers and they appear pretty robust & I intend using them soon on my D.I.Y. attempt! :shock:
Great thread this along with Opcode's blade runner :wink: - keep it comin guys........
Emidisc

User avatar
EpicenterBryan
Posts: 738
Joined: Sun Aug 10, 2014 9:01 pm
Location: Eugene, OR USA

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 31999Unread post EpicenterBryan
Sat Oct 25, 2014 12:34 am

Zero time left tonight, but expect a big update this weekend. Replacement of the ABS plastic / Nylon stylus holder was the key. Check out that bitching stylus holder! I made that from a chunk of 1/2" x 1/2" bar stock! It only added 1 gram compared to the plastic version.
IMG_2997.JPG
It took several nights of messing around and failures... and a 3D printed jig to set the angles in the vise right for drilling and tapping.

Stay tuned for the full update this weekend....


Bryan
You do not have the required permissions to view the files attached to this post.

User avatar
emidisc
Posts: 220
Joined: Mon Jun 07, 2010 7:57 pm
Location: lancashire

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 32002Unread post emidisc
Sat Oct 25, 2014 9:20 am

Bryan
I reckon the stylus holder/torque tube is the most difficult part of a DIY cutter head and I guess you would agree!
You need to bring together 4 different elements
:- Stylus,
2 x driver links,
torque tube support,
Stylus screw,
And all need to be done accurately and yet still allow free movement of the stylus
I have been thinking along the same lines regarding a fixture were all the elements are brought together and clamped at the correct geometry and then soldered,
What I have noticed Bryan with these drivers is due to the back of the unit being open it is easy for debris to to get in there I have ended up closing this with some pvc tape for now!
Looking good can't wait for the update........

User avatar
markrob
Posts: 1623
Joined: Wed Nov 28, 2007 1:14 am
Location: Philadelphia Area

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 32004Unread post markrob
Sat Oct 25, 2014 5:21 pm

Hi,

That seems like a big hunk of metal! 1 gram is quite a bit of extra mass to sling around. Looks like you can remove a good deal of the mass with some additional machining. If you calculate the peak g force required to accelerate a 1 gram mass moving sinusoidally at 5 cm/sec peak velocity at 20 Khz, it gets pretty scary. This is in addition to all of the other masses in system.

Mark

User avatar
opcode66
Posts: 2700
Joined: Sun Nov 08, 2009 10:56 pm
Contact:

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 32007Unread post opcode66
Sat Oct 25, 2014 8:03 pm

The entire moving mass for both channels and the torque tube in my head are about 2 grams. I think less actually. I just reduced the size of every part.
Cutting, Inventing & Innovating
Groove Graphics, VMS Halfnuts, MIDI Automation, Professional Stereo Feedback Cutterheads, and Pesto 1-D Cutterhead Clones
Cutterhead Repair: Recoiling, Cleaning, Cloning of Screws, Dampers & More
http://mantra.audio

User avatar
EpicenterBryan
Posts: 738
Joined: Sun Aug 10, 2014 9:01 pm
Location: Eugene, OR USA

Re: V1.0 of Bryan's 3D printed head with feedback

Post: # 32021Unread post EpicenterBryan
Tue Oct 28, 2014 1:01 am

markrob wrote:That seems like a big hunk of metal! 1 gram is quite a bit of extra mass to sling around.
Wow, what a set of discoveries and failures this last week+.

I had planned to do an update this weekend but all hell broke loose.
Before I share failures, here is an update on the stylus holder shown in the other photo. That one was 2.0 grams, and the plastic one (with the nylon hex inset) was 1.35 grams.

Tonight I got the aluminum stylus holder down to 0.61 grams (from 2.0). I might be able to trim another 0.2 grams but am note sure... The limit on this design is that I've got 3mm thread holes for the push-rods, torque tube connection, stylus set screw and such. They all need threads. I may have to rethink this and eliminate all the threads and glue stuff. It's also at a point where I need to file it by hand.

I did some tests last week with the plastic holder, and tried steel push-rods just to see if the nylon push-rods were attenuating HF transfer to the stylus holder or not. The frequency response was nearly identical even after changes to the torque tube setup. So I thought it might be that the plastic holder was the issue...(It could also be the plastic discs glued on the drivers) So I tried the aluminum holder (and still used the steel push-rods) and it looked more promising in that the frequency response fell off at a much more linear slope, but dropped like a stone. So moving mass is totally an issue like Markrob has always said. It is interesting to note though that goofy response is more of an issue using the plastic stylus holder. That is worth the price of admission. The stylus holder is the most difficult thing to make in this DYI design, and for that exact reason I really wanted to make as many of these parts on my 3D printer as possible. I don't think a usable 3d printed head is feasible unless a LOW-FI result is the intended result.

Now for the failures (oh, there is more to this update?). Last week I did the total rookie move and forgot to engage the overhead and started cutting. By the time I realized what the screeching sound was I toasted a stylus in a locked groove all the way to the aluminum substrate. Yes, you can laugh here with me! I guess everyone has done this once. But it gets worse... A few days later my digital scale arrived and I started to make some cutting force measurements - Since this head design has a fairly giving torque tube, cutting down force is an issue. It has to be light enough, or the whole 45 degree V goes up too much and the geometry of the drivers tweak such that the coils will rub causing unwanted results. Anyway, some how while measuring down force I managed to chip my last good stylus. I didn't know what was going on playback until I had cut far too many tests messing with tracking force, and adjusting cut angle and such.

Some where in there I configured the head with the new aluminum stylus holder and applied locktite to all the connections. When I decided to take it apart again to try to reduce the mass of the holder, the plastic discs on the drivers stripped out rendering them useless in this application. I have 4 spare drivers with discs already installed, but the discs were a previous design and weigh more than the most recent driver discs. Now the game is to reduce mass, so I don't want to use them right now. I have new drivers on order.

To make matters worse this week, my vacuum pump motor started to throw sparks and crapped out Friday night. I found a motor on Craigslist for $10 with the same shaft diameter but it was 1725rpm. The one on the pump was 950 RPM and ones on E-bay are too expensive. I picked up the 1725rpm one locally on Saturday, and discovered it also ran backwards compared to the original motor. After a full day of messing around, I made it work and it was interesting enough I shot a video that I will post. In the end, the pump runs as the same speed as it did before.

Now, the final update... I received aluminum 3mm threaded screws today to use for the new push-rods. I also ordered aluminum discs to glue on the new drivers in place of plastic disc I have used to date. I also ordered 3mm tubing to use as push-rods with no threads to be glued. Not sure how that will work but at this point I'm willing to try anything to reduce mass to increase HF response.

I called Transco today to order new styli, and only got voice-mail. Holding pattern for now.

That's it for tonight.

Bryan

Post Reply