DIY JE-990 DOA

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Damn, I guess the smarter fellas aren't going to rescue me just yet...


I suddenly realized that my second statement above was probably more likely to be misunderstood than the first. So just to clarify, my very selfish reason for doing this 990 layout is so that I can build a pair of mic preamps in the Jensen tradition for my own enjoyment. It didn't seem to me to be enough of a challenge to buy a pair of JH or TSS 990s and stick them in a box with some transformers and the usual pad, polarity and phantom switches. I mean, what's the fun in that?

Some more thoughts about the inductor: it seems to me that one of the major goals of the unique construction of the original is meant to minimize the DC resistance so that at audible frequencies the noise voltage level at the differential stage is also at the minimum. That is the basis of the patent.

The other major consideration is that the resonant frequency of the inductor be kept high, which is why we want to keep the shunt capacitance low - to keep the resonant frequency above ft or the unity gain bandwidth of the amplifier (oops, no subscript bbcode on this forum?) and prevent instability.

But if we choose an inductor with a resonance that is above this 10 MHz figure even if it doesn't give us the 0.03 ohm resistance of the original we will still gain a benefit over not using the inductor at all, without creating any instability in the amplifier? That was a question, by the way.

If so, then there seems to be a few likely candidates that will avoid having to wind your own and find this elusive "nickle-compound" core. I mean, I don't want have to start making my own capacitors and resistors to build this thing.

Man, where has Samuel been?

Also, someone sent me a picture of the inside of the JH990. I'm very glad that I hadn't seen them before doing my own layout and it's very clear that his was not used in the making of this Group-DIY layout. I was afraid that since they were based on exactly the same circuit they would end up looking very close but as far as I can tell they are, as they say, chalk and cheese :cool:

Edit: Upon further examination I see that the JH boards sit component side down, with long pins extending through the potting material. A very clever way to get the NPN and PNP output transistors going to the correct power pins without flipping the layout (see the other thread for an earlier version of my board with a wire link). Also, the photos I was sent were of a 990C, (which runs on multiple voltages without changing any components) not a straight JE-990 so there are a few more components on his board. That accounts for the 1/8 watt resistors and an SMD - there's barely enough room with 1/4W and no SMDs on my boards.

Ever feel like you're talking to yourself?
 
Man, where has Samuel been?
On holiday! :green:
After handing in my master thesis I needed one week off but now I'm back for new doings!

But if we choose an inductor with a resonance that is above this 10 MHz figure even if it doesn't give us the 0.03 ohm resistance of the original we will still gain a benefit over not using the inductor at all, without creating any instability in the amplifier?
Yes. An above-zero DC resistance of the ferrite bead/inductor increases voltage noise, but anything below 5 ohm is really insignificant in practice. A high resonance frequency doesn't hurt at all.

However I'm not at all sure that the inductor really needs a resonance frequency at/above 10 MHz. The impedance is paralleled by the resistors, and thus the impedance of the inductor needs to fall pretty low (say to 100 ohm) till it gets significant. i.e. it lowers stability.

The 20 uH figure is of some importance though, as lower inductance would move the first pole in the response up in frequency.

I'd say reasonable specs for the inductors are 15 to 25 uH, DC resistance below 5 ohm and resonance frequency above 5 MHz. Note the Schaffner part I posted in the other thread.

Samuel
 
Hope you had a nice holiday! Don't ever do it again! :razz:

At first I hadn't really noticed you were referring to Schaffner in the other thread and Fair-Rite in this one...because the numbers are so similar:

2943777741
2944666631

Is that just a coincidence or is there some significance to the numbers?

And why do you refer to impedance of the material in the other thread? From my meager understanding I would have thought the Fair-rite "73" material was the right stuff, having the permeability figure referred to in the Jensen report.

Anyway, if you're right - and I have no reason to doubt that you are - we're in luck, because Mouser has 121 in stock and 3800 on order for 30 cents in single quantities. Good luck for everyone but me, since I received an order from them just last week...

Going completely off-topic again - a friend of mine, when questioned about why he was talking to himself, would say "I'm not talking. I'm listening!
 
I think Fair-Rite is part of the Schaffner group, so I guess all Schaffner beads are actually Fair-Rite's or the other way round.

From my meager understanding I would have thought the Fair-rite "73" material was the right stuff, having the permeability figure referred to in the Jensen report.
I didn't check the material, I just looked for a ferrite bead meeting the specs I mentioned above. 2944666631 is allready assembled, so it is possible to refer to an impedance curve. Where does the Jensen report refer to the permeability figure?

Unfortunately we don't have an impedance plot of the original Jensen inductur available; so the only way to check if my suggestion is a good one is to build one and check transient response...

Samuel
 
[quote author="Samuel Groner"]The 20 uH figure is of some importance though, as lower inductance would move the first pole in the response up in frequency.

I'd say reasonable specs for the inductors are 15 to 25 uH, DC resistance below 5 ohm and resonance frequency above 5 MHz. Note the Schaffner part I posted in the other thread.

Samuel[/quote]

Readily available from RS are these Toko's:

C228450-01.jpg


22uH

RS-stock#:214-6237

? 0,58 excl. VAT

the second from right on the pic

max Rdc: 0·08 Ohm

res.freq. not stated, but since the 100uH coil has a stated res.freq of 6.1MHz I guess the 22uH one will be fine.

A general purpose RF chokes suitable for power decoupling in logic circuits, IF tuned circuit applications and filters etc. The low Q higher current inductors are especially suited to low-medium power d.c.-d.c. converter applications.
Open ferrite bobbin construction.
Leads are on a standard 5 mm pitch and are 0·7 mm diameter.

Would these do or am I missing something ? Would these fit physically ?
 
Where does the Jensen report refer to the permeability figure?
Section 7.2 Emitter Inductor Construction.

I'm going to place an order for those stocked Fair-Rite parts and give them a try.

The other night at school I took the plastic covering off a 33uH inductor (that was as close as they had) hooked it up in series to a resistor, fed it some 225kHz and watched it on the oscilloscope while pulling off windings until it measured 20uH.
33uH-20uH.jpg

This one has a smaller footprint, but the board is made for a lead pitch of around 5mm and there is enough space in there to fit the originally specified part which is around 6.1mm so those Tokos should drop right in.
JE-990_prototype_bobbin_inductor.jpg

Anyone doing a board run in Europe or is Brasil the only place doing them so far?
 
I've been watching this avidly and am constatly amazed by the nuts that are cracked! This 990 de-mystification/ conquering is awesome!
I could get some 990s made up if people want them in Europe. I'll definately have some but don't really want to get 100 if there's no others...I can't believe there are no others :green: :thumb:
Also I don't want to clash with anyone else so speak now or forever hold yer peace :grin:

chef
 
Nice, so that kind of coil seems usable. Did you have some time for listening yet ? Or could compare this type of coil to those you showed before ?

I think I better add some of those 22uH Toko's to my next RS-order ! :wink:

Sure, would be interested in some PCBs for that, say 2 or 4 (not going to build complete desks with it).

Anyone knows of a good source for the LM394 here in NL ?
 
Go Chef!

I'm pretty sure he means PCBs - no one would consider getting into the 990 business now except maybe Behringer :wink: Seriously though, there are at least two, maybe three businesses doing finished modules already who have earned their place in the audio world and deserve to be promoted and supported. To me, this is just an interesting distraction and a fun way to share some DIY.
Nice, so that kind of coil seems usable.
Looks like it according to what Samuel has written. It would be great if there were an easy to get off-the-shelf part for this circuit.
Did you have some time for listening yet ?
Err...you talking to me? My idea is that other people will build these things and do the serious listening and measuring. The pcb design went OK, the building of the prototypes went fine but after that I'm out of my league!
 
[quote author="mnats"]To me, this is just an interesting distraction and a fun way to share some DIY.
Nice, so that kind of coil seems usable.
Looks like it according to what Samuel has written. It would be great if there were an easy to get off-the-shelf part for this circuit.[/quote]
Seems like the Toko does the trick then yes, and it's cheap. That six-hole core with the special edition patent style wiring looks cool though :thumb:

Did you have some time for listening yet ?
Err...you talking to me? My idea is that other people will build these things and do the serious listening and measuring. The pcb design went OK, the building of the prototypes went fine but after that I'm out of my league!
I see :wink: :thumb:
I understood you both have a six-hole core Mnats990 and a Toko-style Mnats990, right ? Was just curiuous if you'd had put an ear to those for a first impression.
 
I understood you both have a six-hole core Mnats990 and a Toko-style Mnats990, right ? Was just curiuous if you'd had put an ear to those for a first impression.
Naw, I'm just fooling around with the inductors until I decide what is going to go in there. The six-hole core I have probably won't go in because it just looks wrong with some windings doubled up in two of the holes. All I've measured with the bobbin-style is the inductance; I have no idea how I'm going to measure the other parameters with the limited equipment I have.

I was thinking the other night that since I've got enough parts here I might just put together a headphone amp with two of these babies. That way there would be nothing downstream but a pair of K240s so all I'd have to take into consideration would be whatever source I was driving it with. Whenever someone writes about this piece of gear or that component sounding sooooo good I always wonder - what are they feeding it with and what is it feeding?

Seems like the Toke does the trick
Hey, good idea!
 
[quote author="radiance"][quote author="Swedish Chef"]I could get some 990s made up if people want them in Europe.
chef[/quote]

You mean PCB's?[/quote]

Ya sorry; PCBs! :green:

chef
 
[quote author="mnats"]Go Chef!

I'm pretty sure he means PCBs - no one would consider getting into the 990 business now except maybe Behringer :wink: Seriously though, there are at least two, maybe three businesses doing finished modules already who have earned their place in the audio world and deserve to be promoted and supported. To me, this is just an interesting distraction and a fun way to share some DIY.
[/quote]

Yeah I do not have Beh-spit icon here-ringer allusions. :green: Pcbs was what I was thinking, but was unclear. sorry for any confusion casued by my fatigued brain!

I'll talk to my fabrication folks on Monday and let you all know what happens asap :grin:
At some point I will find the headspace to go through all the cool stuff in this thread and try and see if I can't get some to stick :sad:

Cheers

chef
 
Well, woke up this morning, made a cup of Joe, went to check my email and found a message from a Mr. Hathaway which said in part: "My apologies for the delay in responding to you. We still have the inductors available but not the threaded assemblies."

If you look at the bottom of the Jensen price list, the part number is FB-6 (referred to as "JT-FB-6" in his email) - bag of 100 for six bucks.





:grin:
 
Cool, that's good news! I allways wondered what's the difference between the FB-2 and FB-6 but never got around writing them an e-mail.

If Jensen orders in Europe wouldn't have these horrible handling and shipping fees...

Samuel
 
Hi Mnats, I saw your pdfs and gerbers files and there an error on this.
There are no CR11 placed.

I think CR7 is the place for CR11, then CR8 is the place for CR7+CR8.

Plz check your designs and let me know if Im wrong or not.
 
Back
Top