Project "Groove Scribe"- A DYI stereo 45/45 head.

[EpicenterBryan]

I have 2 new drivers on order to salvage new coils. Until they get here, I want to try a few things but have to live with what I have that's already glued. I have one driver that has a carbon fiber disc already glued to a Ninja Flex diaphragm, and the carbon fiber is already glued to the coil but I haven't installed a push rod or coil. So I'm going to try something quick to see if this cone idea is significantly better.

IMG_4334.JPG

I'm gluing a cone on the only surface I can just for a quick test. It's on top of the NinjaFlex. But the push rod will not connect to the carbon fiber disc that is still present, it will only attach to the cone. I have to work with what I have right now...

I have made several cones, and glued one to the driver but not the push rod / coil shown. Before I glue the push rod I have to make a new spacer for the final driver, and also a new alignment tool.

IMG_4333.JPG

And before I spend that time, I want to see if CA will actually bond the cone to the NinjaFlex. I know the black stuff works but takes days to really stiffen up. So I also have a cone glued to a diaphragm to set up over night. I'll see if that holds up before I spend time on new spacers and alignment jigs this weekend.

IMG_4335.JPG

If any of this stuff works out, I have a few tricks I've learned that I'll pass on.
I should also say that what I'm really looking for is a big improvement in the phase plot which I eliminated from what I posted the other day. Right now, I don't care as much if the HF response is better. I have a few other things in mind that should help. And if in the next round - the carbon fiber disc goes away - that also reduces some mass.

Wish me luck this weekend.
And to all the Trolls on the east coast, be safe out there. This storm looks really nasty.

Bryan

[cosmO_O]

Hi Bryan!

Your thread is very inspiring, at the moment i try to print usable cones... You ever try to print some or there is a special reason about the aluminium ?
Thanks for the answer!

Lali

[Snug Music]

wow,..Bryan,..thats awesome!!!
great tunes u made!
i hope thats it!!! this look´s good for right high-end of these driver!!!

greetings Scotty..

we belive in u Bryan...

[Snug Music]

Hi Bryan!

Your thread is very inspiring, at the moment i try to print usable cones... You ever try to print some or there is a special reason about the aluminium ?
Thanks for the answer!

Lali

Hilali, ..
I think it is related to the transmission through the aluminum!
Also, the cone and the shape of which, has a direct influence on the overall result.
Makes itself felt immensely in the heights!
greeting Scotty

[Snug Music]

Hey Bryan, ..
I have found the error with me hehe ... ..
have a Hollywood Nightline 2.0 1200W amplifier, as I now have screwed him .. Peng, .... there's nothing in it (a small board with a few components and a small toroidal transformer ... nothing else!) This may never bring the required performance! Have now taken the Peavey cs-400 PA-Amp and lo and behold! it works as it should! hehee, ... that's it!

Scotty

ps: now, i will make tests...

[EpicenterBryan]

Great way to start the day.
The CA glue test is a go!

Left: It actually holds up. Right: It even bounces back.

CA glue test.jpg

So now it's time to make a jig and spacer and get on with a new test.

Bryan

[EpicenterBryan]

You ever try to print some or there is a special reason about the aluminium ?

Hi Lali,
yes, my goal was to make it as strong and a light as possible and make sure it didn't suffer from compression issues that could cause the bulk of the phase issues in HF that I'm running into. This is just a test to see how much better it is. I still expect issues because it's glued to the diaphragm (not the driver directly) but that's the best I can do until I can start over with a fresh coil.

Hey Bryan, .. I have found the error with me hehe ... ..

That's odd. I looked at the specs on that amp at it looks reasonable. But if you are getting better results with a different amp I guess it's possible the amp needs repair.

Bryan

[Snug Music]

Hey Bryan, ..
have just seen that the membrane of you is quite flexible. have been wondering why you have a disk and not a spider as ciuens and I. aha, that was the reason. Therefore, my question to you, how flexible your diaphragm ... hehee ... I'm already a little pain in the ass ... lool

regards Scotty

[EpicenterBryan]

Therefore, my question to you, how flexible your diaphragm ...

This might help:
https://www.youtube.com/watch?v=gTIFs_ILu1w

Bryan

[Snug Music]

Therefore, my question to you, how flexible your diaphragm ...

This might help:
https://www.youtube.com/watch?v=gTIFs_ILu1w

Bryan

yeah,..thanks Bryan...i never seen before..
hmm .. since I could have come on it himself.
had probably my head full of unimportant things ...
But know, who can read is a clear advantage! hehee lol

best regards Scotty

[EpicenterBryan]

The quick cone test data is in!
I am speechless. The frequency response looks like crap, but the phase relationship is so much better than anything I have messed with since I started taking transfer function data. And this is the first time the "delay finder" function actually calculated a 0.0ms system latency. I hope Mark has some time to make a quick comment about this. If not, I'll give it a shot after I do another quick test tomorrow...

Quick Cone Test.JPG

Right now a standard feedback loop could not be closed without some compensation. Once the phase relationship goes past about 140 degrees the loop could go unstable without lead or lag as needed to keep within that range (40 degrees of phase is a good amount to hold in reserve just in case - below 180 degrees which is the theoretical limit).

Remember, this was only a test. It had the cone glued on top of the NinjaFlex diaphragm. So there are several things that need to change now that I see how much better the phase plot is. In this test the coil attached to a carbon fiber disc, which then attached to a NinjaFlex diaphragm, which then attached to a beer can cone, which then attached to a push rod that has the coil wound on it.

The next version will have the cone attached directly to the coil. The push rod (different and lighter ) attached to the cone. The NinjaFlex will be thinner and will attach to the edge of the cone.

Damn. That is so much closer to a range of phase change (and frequency response) that could be compensated with feedback.

I'm thrilled. Thanks again to everyone who has shared what they have been messing with. You guys nailed it. The phase issues came from the transition from the large coil diameter to the smaller push rod diameter via what ever shape or material was used.

Bryan

[markrob]

Hi Bryan,

Nice work on forming the cone. I learned a few tricks watching this come together.

Phase response looks pretty good. You might even be able to close the loop around this extra response bump since I don't see much additional phase shift beyond -90 degrees.

I wonder if the second resonance is from the aluminum cone. Maybe try some damping on the cone surface ala' the thread on the reverse engineering of Souri's driver. I used paper for my cones to avoid the ringing of metal. It seems better damped without any treatment (other than the lacquer I used to stiffen).

Mark

[Snug Music]

The quick cone test data is in!
I am speechless. The frequency response looks like crap, but the phase relationship is so much better than anything I have messed with since I started taking transfer function data. And this is the first time the "delay finder" function actually calculated a 0.0ms system latency. I hope Mark has some time to make a quick comment about this. If not, I'll give it a shot after I do another quick test tomorrow...

Quick Cone Test.JPG

Right now a standard feedback loop could not be closed without some compensation. Once the phase relationship goes past about 140 degrees the loop could go unstable without lead or lag as needed to keep within that range (40 degrees of phase is a good amount to hold in reserve just in case - below 180 degrees which is the theoretical limit).

Remember, this was only a test. It had the cone glued on top of the NinjaFlex diaphragm. So there are several things that need to change now that I see how much better the phase plot is. In this test the coil attached to a carbon fiber disc, which then attached to a NinjaFlex diaphragm, which then attached to a beer can cone, which then attached to a push rod that has the coil wound on it.

The next version will have the cone attached directly to the coil. The push rod (different and lighter ) attached to the cone. The NinjaFlex will be thinner and will attach to the edge of the cone.

Damn. That is so much closer to a range of phase change (and frequency response) that could be compensated with feedback.

I'm thrilled. Thanks again to everyone who has shared what they have been messing with. You guys nailed it. The phase issues came from the transition from the large coil diameter to the smaller push rod diameter via what ever shape or material was used.

Bryan

Hey Bryan,..

That looks great,.. im afraid u have make this.......and yes the better phase comes from the cone on rod.
I think the crapy frequenzy can u change, when the voicecoil glued direktly on the cone, without the carbonedisk!
I stay tuned...
Best regards Scotty

[Snug Music]

Hi Bryan,

Nice work on forming the cone. I learned a few tricks watching this come together.

Phase response looks pretty good. You might even be able to close the loop around this extra response bump since I don't see much additional phase shift beyond -90 degrees.

I wonder if the second resonance is from the aluminum cone. Maybe try some damping on the cone surface ala' the thread on the reverse engineering of Souri's driver. I used paper for my cones to avoid the ringing of metal. It seems better damped without any treatment (other than the lacquer I used to stiffen).

Mark

Yes i forggot to say...
Take a little liquid rubber and make this around in distance of 10mm from middle on the Aluminiumcone.
I have do this too!
Yes, i have seen this by souris driver this blocked the metallringing!

Scotty

[Snug Music]

https://www.youtube.com/watch?v=zApmGFDA6ow

3d print metall

[Snug Music]

http://bindulin-shop.de/de/produkt/spachtelmassen/1-komp_spachtel/bindulin_fluessig-aluminium_60g_tube__SPA10.html

I think this can help.

[Snug Music]

i think i give up with this Transducer!
To much frequenzy-shit...

[EpicenterBryan]

Sunday night I tried to close the loop with Caruso Massimo based on the driver configuration from the last frequency response post. I wasn't sure if there was enough phase margin but what the heck. Sure enough, it wanted to oscillate when the FB gain was brought up. Mark suggested trying to dampen the cone with something, so I put some RTV on it - and let it dry over night and again, it wanted to oscillate - each time around 3600hz. I didn't have access to the inside of the cone, just the outside because the carbon fiber disc was still installed.

I also noticed Caruso Massimo added an additional 45 degrees of positive phase shift in low frequencies. Not sure where that is coming from and will have to look over all my settings on the board again. Perhaps my AC coupled VU drivers are involved. My notes on how I configured the Caruso boards are:
R606 270 ohm for about 35db gain on FB amp.
C401 (forward comp 2nd response) REMOVED.
C405 REMOVED (Caruso)
R408 10K (Caruso)
C403 REMOVED (forward comp)

Also tried with R405 (bass forward) installed and also shorted. The big difference was if shorted, there was not enough gain.

Here is the response with RTV on the cone right before breaking into oscillation. The good news is that it looks like the RTV did help settle down the upper frequency bump.

IMG_0100.JPG

Since the cone is 10mm tall, I had to add a 10mm plastic spacer to make it fit with the other stuff I've built and I'm not sure if my magnetic shield is working properly now. So I'm going to have to check that. Perhaps my spacer will need to be steel. Perhaps I can wrap it in metal. I'm going to try taking a driver structure, break out the magnets, Install everything else and run signal into the driver coil. If I get anything on my feedback coil then it's BS signal from the drive coil getting into the FB coil.

For the next round, I had to make new cones and new diaphragms. This time, I made the base of the cones 2mm wider so I could hand form "hat brims" on them so I had a nice solid flat surface to glue it to the driver coil, and also a nice flat area to glue the diaphragms to the cones. I experimented with dipping a few cones in "Plasti Dip" (2 coats) both inside and out. I've now got a feel for how to do this. I also printed new diaphragms in 1/2 the thickness as the last ones since this can go directly on the cone now. I'll have to made several alignment jigs now to keep all these parts centered in each step.

I also decided to see what the implications were on moving mass...

Top left, a bare cone. Top right, after 2 coats of "Plasti Dip". Bottom, what the carbon fiber disc weight was that the cone totally replaces...

Cone - coated - carbon.jpg

That's about it for the night.

Bryan

[EpicenterBryan]

i think i give up with this Transducer!

Give up when we have the rest of the data. I have new drivers on order, but was able to salvage (yet again) the coil from one and I hope to have data on a new configuration this weekend.

If it were easy, we all could buy a stereo head for $500 with feedback. As you now know, there are many hours involved trying to make one yourself since we can't just go out and buy it. I still want my Caruso!

Bryan

[Snug Music]

i think i give up with this Transducer!

Give up when we have the rest of the data. I have new drivers on order, but was able to salvage (yet again) the coil from one and I hope to have data on a new configuration this weekend.

If it were easy, we all could buy a stereo head for $500 with feedback. As you now know, there are many hours involved trying to make one yourself since we can't just go out and buy it. I still want my Caruso!

Bryan

hey Bryan,..
OK, I think .. you're right! will wait ... I'm an optimistic person ... hehee ...
am looking forward to your update!

best regards ...
Scotty