• These commercial threads are for private transactions. diyAudio.com provides these forums for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members, use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

GB For RMI-FC100 Parts

Status
Not open for further replies.
Jump to Latest
#82
pooge said:
I'm still unclear exactly how R-cores are wound. They still appear to have secondaries wound over the primaries. If so, I don't understand how that prevents capacitive coupling like e-i cores.
Click to expand...

Honestly spoken I do not either. I thought they were wounded primary one side, secondary other side.
😕

Ernst
 
#83
I assumed you inquired at Custom Magnetics in the US for the R-cores, also made in India.
I've got a private supplier for Japanese made R-cores, dollar-wise not interesting though.
The Shilchar company in India, that manufactures the French transformers, lists separate bobbins and cores for R-cores.
It would be interesting to know what the cost of bobbins/cores would be, and the minimum order number.

Primary on both sides and CT connected, secondaries separate on each bobbin.
 
#85
From what I gather, the advantages of R-cores is their efficiency due to the round core; cooling, due to the fact that the windings do not cover the entire core; and low stray field due to split windings oppositely oriented. Marketing primarily compares them with e-i types and not toroids, but they also apparently have lower input current surges at turn on than toroids. It still appears to me that the secondaries are wound over the primaries, and the electrostatic coupling is stopped by an electrostatic shield. Marketing diagrams are still vague on this. Toroids can have the same shield.

All this is well and good, but at what cost and availability? Custom Magnetics is a bit snotty and seemingly not too accommodating to the little guy. They would also be cost prohibitive. They are known for not answering email inquiries, even if their minimum order of $500 (or was it $750?) is exceeded. They are the only supplier in US. The other one bailed out due to low demand, and issues with foreign manufacturer.

I'm still waiting for a response from the R-core manufacturer, but I'm just parallel processing a possible toroid alternative.
 
#86
R-core vs. toroid

Ok, thanks for some clarification. Seems to me a if a fine toroid with shieldwindings could be close to an R-core design.

So there are a few options:

1: Buy custom made R-Core from for east >30 € due to transport of just a few parts
2: Buy french (indian) R-Core and modify it on my own after Mihais proposal ~ 20 + ?? €
3: Buy 4 standard 2x18V ~ 10 VA toroids ??€
4: Buy 4 standard french 14VA 2x18V R-cores: >60€
5: Get custom made toroid design with shieldwindings. ?? €
6: Buy German custom made E-I design >60 €

Don't think that enough people were interested to ask for a few in India.

So option 2 seems to be the cheapest and most DIYest 🙂
Thinking about it, if I could manage this. Probably I give it a try.

Thanks for all your input 🙂

Ernst
 
#87
Having standard stock is a bonus. Wish I had your options.

Make sure you check for regulation. My initial inquiry on R-cores found the standard offer to have 10% regulation. While not a problem for the input stage, due to its regulation and fairly constant current, it is a bigger issue for the output stage. Part of the hassle I am dealing with is communicating the specs for better regulation. It requires a bigger core and fatter wire. You can put less fatter wire on a given core size. Therefore, changing to the larger gauge wire means less VA for a given core. Going through these iterations with a middle man is not a walk in the park.

Alternatively, communicating with a toroid manafacterer directly about these things has been a pleasure, relatively speaking.
 
#89
Mihai,

If my French is what I think it to be, this is the type of marketing crap I see associated with R-cores. There is no inherent reason an R-core is better than a toroid in this regard, as each can be good or bad. Don't get me wrong, I would take an R-core over a toroid for various reasons if I could, but I have easily found 4% regulation in a toroid (and found 10% on an R-core). The value got a little bit worse when I specified bifilar windings, but it seemed a good tradeoff.

I just wish the marketeers would cut out the hype and stick to promoting the actual advantages from the stand point of inherent advantages vs. how well they are done.

I remain envious of your options.
 
#90
NJL 0302/0281

Guys,

I try to organize a GB for these devices. Bad thing is, Digikey has got a minimum order by 225 parts each :-( Lets see, if we can exceed this number.

I'd sell them for the same money plus fee for packing, and shipping.

Part price: see here:
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=NJL0302DG-ND

I try to organize matching, even I do not know so far, how to do.
How much this would be then, I can't say of course. At least, it will be a lot of work ;-)

If you are interested in, modify the WIKI page - therefor see my signature.


Cheers Ernst
 
#92
Andrew,

I would order the recommended ones. Since minimum order is 225 each, I think there is enough room for matching and getting the right ones for all.
I for myself have 15 each, but have not done the matching so far, but I'm think about starting soon.

The more people took part, the higher the probability for good matching :clown:

Cheers Ernst
 
#94
Hi Folks, Please, don't care to much for output transistor matching.
In pratice, these modern transistors have too little spread, when compared with old school transistors.
In matter of reliability, please think about this:
One pair is what is necessary in pratice, for this level of power.
Because this, there is no need for tight matched current sharing between 3 pairs.
The beta matching is of little significance, because the transistors will be used as VOLTAGE FOLLOWER, with plenty of driving current, and,
if 10 millivolts of spread between 3 pairs appear to be so much, remember this is only at quiescent current, but when each transistor is conducting, say, 1A, the spread of 10mV represents only 10% of unbalance in the current sharing, for 0R1 emitter resistors.
For sceptical people, I want to say I'm involved with amplifiers since 1980, both pro-sound and hi-fi.

cheers,
Marcos
 
#95
NJL 0302/0281

Just tried my samples with an ordinary multimeter, quick and dirty.

Mine 15 pcs 0302 are all around hFe 117 (+0/-3)
mine 15 pcs 0281 are all around hFe 80 (+0/-2)

Have not in mind, if matching PNP vs NPN is important? Is it?

Cheers Ernst
 
#96
Re: NJL 0302/0281

ernesternest said:
Just tried my samples with an ordinary multimeter, quick and dirty.

Mine 15 pcs 0302 are all around hFe 117 (+0/-3)
mine 15 pcs 0281 are all around hFe 80 (+0/-2)
Click to expand...

Ernest,

An DMM have 1mA Ic current for hfe testing, which is to little for power transistors.
 
#97
matching BJTs

Hi,
do not use an ordinary DMM hFE tester.
As far as I know they all test by holding the base current constant.
They then display the collector current.
This does not measure the Vbe and does not hold the device at a known temperature.

You must measure both Vbe and Ib when the Vce is held constant and the Ic is held constant (or at a range if fixed settings).
Fixing both Ic and Vce fixes the dissipation and in turn fixes the temperature.

This needs a low voltage DC supply, a CCS, a variable resistor and a calibrated resistor.

For devices in parallel the Vbe matching is very important. Don't just match hFE @ bias current.

BTW,
I got a 30% spread in hFE in the NPN devices and 10% spread in the PNP devices. Vbe spread was just 10mV for both groups.
 
#100
Capacitors

Hi,

capacitors have arrived from Digikey 🙂
Will forward them after Karneval :clown:

Thank for advices regarding matching Onsemis.

Maybe I read it month ago, but if, I have forgotten. Is it necessary to match PNP vs NPN devices as well?

Seems that a transistor GB wount happen....

Cheers Ernst
 
Status
Not open for further replies.
Top Bottom