The fake MF resistor saga

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: Kent, luketeaford, Joe.

User avatar
raisinbag
Super Deluxe Wiggler
Posts: 1708
Joined: Thu Aug 18, 2011 11:40 pm

The fake MF resistor saga

Post by raisinbag » Thu Aug 23, 2012 9:09 am

Ok so recently after reading some of the posts about fake resistors I tested a few of the ones I bought from tayda. I just wanted to make sure that the large lot of assorted resistors I bought were real. When I did the "burn test" my tayda resistors dropped from 100kohm to 94ish and when I tested another one from another source it only dropped to aprox 98kohm. So I contacted tayda to ask them if I had fake resistors. They actualy responded said that they would Investigate this. I have been hapy with everything else i got from them so this is not a flame post, just trying to figure out the truth and figure i will get an honest unbiased answer here. Here is response after investigation:

Edited out names etc.

Tayda person said
"I checked with our QC people and also with manufaturer, it is not possible to manufacture carbon film resistor look like metal resistor. And their price difference is so little I don't think anybody would bother to manufacture metal film a look like carbon film capacitor.

Also, we sell hundreds of thousands of these resistors every month and we never had any problem report from buyers for this product category.

If you find any other information please let us know so we can do further investigation."

End Tayda.

So can anyone really tell me the scoop? Do I have fakes and this company is either lieing or ignorant? Do I actually have MF resistors with a shitty temperature "coefficient" (if that is the correct term). Wtf do I do wih my investment of resistors? Can I use them for most builds with exception to vco's where I needs stable pitch. I'm kinda pissed off and confussed, as I (relative to me) spent a lot of money and thought I was all set for many many future BOMs. And now I feel like I wasted $ and am back to square one.

frijitz
Super Deluxe Wiggler
Posts: 1658
Joined: Mon Aug 31, 2009 3:37 pm
Location: NM
Contact:

Re: The fake MF resistor saga

Post by frijitz » Thu Aug 23, 2012 9:19 am

raisinbag wrote:So can anyone really tell me the scoop? Do I have fakes and this company is either lieing or ignorant? Do I actually have MF resistors with a shitty temperature "coefficient" (if that is the correct term). Wtf do I do wih my investment of resistors? Can I use them for most builds with exception to vco's where I needs stable pitch. I'm kinda pissed off and confussed, as I (relative to me) spent a lot of money and thought I was all set for many many future BOMs. And now I feel like I wasted $ and am back to square one.
MF resistors come with different temp coefficients. Without knowing how hot you are getting them, it's hard to tell what the coefficient is. You could try measuring a CF resistor at the same conditions -- that would be a better comparison.

Ian

Jarno
Super Deluxe Wiggler
Posts: 2841
Joined: Mon Nov 15, 2010 6:34 am
Location: Rosmalen, NL

Post by Jarno » Thu Aug 23, 2012 9:55 am

Are the temperature coefficients of MF and CF resistors sufficiently different because of inherent material properties, or can the temperature coefficent be "tuned" or is it already similar?

Would be interesting to test this, we do have the equipment available at work (but the test labs are always overloaded of course). Maybe think of a simple test jig with an element with a known temperature coefficient and compare the DUT with that.

User avatar
decaying.sine
Broken <--> Fixed
Posts: 5397
Joined: Wed Jun 03, 2009 10:05 pm
Location: New Haven, CT USA

Post by decaying.sine » Thu Aug 23, 2012 10:44 am

For 1/4w metal film a common TC value would be something like 50ppm/C, which you'd find for the Xicon 271 series resistors.

You need to ask them who is the manufacturer and what is the manufacturer product number for the resistors in the assorted kit you bought. Then track down the datasheet to find the TC.

edit* It looks like the manufacturer might be Royal Ohm. They have 100, 50, and 10ppm/C TC metal film resistors. I am sure it is not 10. If they were really cheap, maybe they were 100's and that is why you are seeing the change.

Like Ian and Jarno were saying though the context is critical. In other words, how hot are they getting. If they are not getting hot at all, i.e., little temp change then that might be a little more troubling.

FWIW, I use almost exclusively Xicon 271 series for 1/4 watt metal film and occasionally Panasonic or KOA for "off values." I've been happy with them over the past few years.
Brian
"I must create a system or be enslaved by another mans; I will not reason and compare: my business is to create." William Blake

"Vactrols ringing, Dude." "Thank you Donny"

User avatar
mechie
Wiggling with Experience
Posts: 334
Joined: Thu Jan 20, 2011 6:36 am
Location: North Lincs, UK

Re: The fake MF resistor saga

Post by mechie » Thu Aug 23, 2012 11:37 am

raisinbag wrote:my tayda resistors dropped from 100kohm to 94ish and when I tested another one from another source it only dropped to aprox 98kohm.
What kind of temperature change were you using for this 'burn test'?
How confident are you that your measurement of resistance was :-
(a) accurate, and
(b) repeatable
I have one multimeter (a Fluke - not a cheap make) that lies like crazy when the battery is low!
Without knowing exact test conditions it is impossible for anyone to reach a proper conclusion.
raisinbag wrote:I'm kinda pissed off and confussed, as I (relative to me) spent a lot of money and thought I was all set for many many future BOMs. And now I feel like I wasted $ and am back to square one.
The figures you quote suggest that these resistors are about 5% from ideal... now the question is...
does it really matter?
Most circuits that are sensitive to such things will have some form of trimmer to calibrate it - a typical capacitor will be 10% and any design must be able to cope with that or forewarn you to get expensive bits!

I don''t see a huge issue here - if you want top-notch performance buy known top notch components (at top notch prices), if, like me, you build for a hobby, then get stuck in and use these resistors!
If a circuit is really critical I would expect the designer to tell me which bits need to be high quality, the rest will be what I have on hand, RadioSpares left-overs, Velleman selection boxes or bits cannibalised from old TV sets :deadbanana:

Don't forget that people lust after synths from the 70's - they didn't have the quality of components we have today!

[/soapbox] ... Ok guys, flame away!

User avatar
mattrick
Common Wiggler
Posts: 188
Joined: Wed Apr 18, 2012 10:16 pm
Location: Madrid, New Mexico, USA

Post by mattrick » Thu Aug 23, 2012 1:06 pm

Don't put naked flames near your synthesizer

:tu:

Seriously though, i'm with mechie on this. I don't think anything will fail to operate due to small variations in resistance. I personally like the idea of some random inconsistency between different examples of the same thing; for me that's a major part of the attraction of analog circuits.

Most of my resistor stock is actually pretty tight, i don't recall the manufacturer right now, but i splashed out on a complete E24 set with 100 of each value some time ago. Recently I topped up my commonly used values (100k, 10k, 1k iirc) with some cheap shit off Ebay, and I measured a bigger variation compared to my original set even without applying heat. I was slightly disappointed at first but then I reminded myself of the above statement and carried on soldering.

I think if you really care about component tolerances, it's best to buy products from a well known manufacturer, from a well established source, and pay the price for that benefit.
san pedro labs | creative audio electronics | custom synth manufacture
my work: http://sanpedrolabs.com
my music: http://www.soundcloud.com/mattrick

User avatar
Memory_Leak
Wiggling with Experience
Posts: 417
Joined: Wed Jan 25, 2012 1:56 pm
Location: 127.0.0.1

Post by Memory_Leak » Thu Aug 23, 2012 2:09 pm

I am very interested to see how this topic turns out :mrgreen:
We bend the truth to accommodate our human weakness.

User avatar
samuraipizzacat29
adadadadadadadsr
Posts: 1064
Joined: Mon Jun 27, 2011 3:04 pm
Location: Allentown, PA, USA

Post by samuraipizzacat29 » Thu Aug 23, 2012 3:37 pm

not that I have anything quantifiable to add to this subject :hihi: but I've decided to switch to koa speer or xicon from mouser for my builds. I don't know if it's legitimate or not, but handling thousands of resistors and the way a "real" one feels compared to what I'm getting from tayda/futurlec is just not worth it. I haven't had one fail AFAIK but I figure if I have only one failure that costs me a chunk of change, it's not worth it vs the $2 I save on the 30-40 resistors per build.
"You can't rule the world by going at it all scattershot." - Pearl Forrester, "The Final Sacrifice"

Always stuff for sale. Pm me, add me on facebook. Send a carrier pigeon. Fox me.

User avatar
xpander
Super Deluxe Wiggler
Posts: 1133
Joined: Tue Dec 30, 2008 6:24 pm
Location: los gatos, california

Post by xpander » Thu Aug 23, 2012 3:53 pm

i just bought a giant metal film resister kit from a chinese ebay seller... hopefully those will fare well. i think a paid about $25 for approximately 1 bajillion.

marvkaye
Wiggling with Experience
Posts: 341
Joined: Sat Feb 04, 2012 1:41 am
Location: Florida

Post by marvkaye » Thu Aug 23, 2012 4:16 pm

xpander wrote:i just bought a giant metal film resister kit from a chinese ebay seller... hopefully those will fare well. i think a paid about $25 for approximately 1 bajillion.
I did the same awhile back, bought 2 kits of 50 each of eleventy different common values for $12 or $13 per kit, with free shipping no less. I notice that they have thinner leads than the Speer or Xicon resistors I get from Mouser. However, I measure each resistor value as I populate my boards (doesn't matter who the supplier is) and of the hundreds of these cheaper resistors that I've used I don't recall a single one being outside the 1% tolerance. (I did have one pack of 33K that was marked 330K, but since they're on tape and their markings matched their values, once I found that they were in the wrong drawer in a mis-marked package they got moved to the right place and haven't caused a single problem.) Now, even though they (the cheaper guys) measure ok I don't use them where values are critical (R/2R ladders, etc) but for everything else I think they're fine. YMMV of course. FWIW.

<marv>

User avatar
daverj
Vintage Video Wiggler
Posts: 8514
Joined: Fri Jun 19, 2009 11:09 am

Post by daverj » Thu Aug 23, 2012 4:16 pm

There's no way of really knowing if the results you got are a problem without knowing how hot the part actually was.

Do you have a 100k 5% resistor that you know is carbon film? How much does it change compared to the other one. If it changes 5-10 times as much, then the other one probably really is MF. The MF ones come in many different temperature versions, so having one MF change 3 times more than a different MF one is reasonable. If it changes as much as a known CF one, then it's a problem, and might not be MF.

User avatar
raisinbag
Super Deluxe Wiggler
Posts: 1708
Joined: Thu Aug 18, 2011 11:40 pm

Post by raisinbag » Thu Aug 23, 2012 4:24 pm

Wow excelent. Some great feedback. I did test the resistor while holding a flame to it and it got bloody hot (how's that for qualitative temp value) I assume WAY hotter than the resistor would ever get, when in operation in a circuit. I did this test because of other threads I have been following here, and the people who wound up with Thier "fakes" seemed to think they had a pile of garbage. I am stuffing a bunch of projects right now (266, RLSeq, timemachine, j3rk stuff and LPGs) and didnt want to create a total cluster fuck if I use these resistors.
After reading your comments I think I will use these. I don't mind a little randomness, as this is the reason in building all this and not using software etc. I will order better quality resistors from now on.

User avatar
J3RK
Super Deluxe Wiggler
Posts: 6770
Joined: Tue Dec 22, 2009 12:46 pm
Location: Seattle

Post by J3RK » Thu Aug 23, 2012 4:52 pm

I have everything from a metric ton of Tayda to Xicon and Vishay. I absolutely love the Vishay (brown/sausage looking) ones. I also like the tiny Panasonic ones that are 1/4W but look about the size of an 1/8W. They're short, and look like little bees. Anyway, I haven't had any issues with any of them, including the Tayda ones. They definitely -feel- cheaper than the others though. (thinner leads mainly) They all test ok at normal temperatures. Not to say they aren't worse off at the extremes though. :despair:

I notice that their tolerances are off a bit more than usual, but not by enough to cause me any concern. If I need something a bit more exact, I go for the Vishays or Xicons.

User avatar
ringstone
antipodean experimentalist
Posts: 2789
Joined: Wed Mar 17, 2010 4:31 am
Location: Trudging through the Capital Wasteland

Post by ringstone » Thu Aug 23, 2012 6:26 pm

Don't forget many vintage designs were made largely with carbon film resistors... and are still prized for their unique sound :hihi:

Cheers
Blair

User avatar
raisinbag
Super Deluxe Wiggler
Posts: 1708
Joined: Thu Aug 18, 2011 11:40 pm

Post by raisinbag » Thu Aug 23, 2012 7:35 pm

Ok cool, the paranoia bubble is slowly deflating. I'm not making modules for sale to stadium playing synth based bands or something. And if there is some minor pitch variation when I play in the desert or arctic, I can live with that. It is honestly for fun and Pleasure. What I really don't want to do is populate a beautiful new 266 set of boards and have them fail drastically because my resistors drift when they are in use. It really took holding the tip of a flame from a lighter to make the resistance drop. I don't plan on burning my modules so I assume that the temp of the resistors will never reach anything like that. For my octave dividers in my miss 10's I used .1% which cost a small fortune. Under normal temps these "questionalbe" resistors were very close to Thier marked values. I will take all that you peeps have said and make sure I research what I buy in the future more closely. So unless someone says explicitly don't build a certain unit with shitty temp coeffiennt MF resistors, I'll just use 'em and say my shit is vintage. ;).

User avatar
Memory_Leak
Wiggling with Experience
Posts: 417
Joined: Wed Jan 25, 2012 1:56 pm
Location: 127.0.0.1

Post by Memory_Leak » Thu Aug 23, 2012 7:36 pm

Just noticed this on the dip-micro site.
100k 1/4W resistor ±1%. They are marked
Metal Film on the box and they are blue
color. We measured thermal stability
around 190ppm/ºC, little high for metal
film, but lower than carbon film.
Btw thanks for all the info on this topic....been very helpful for me :tu:
We bend the truth to accommodate our human weakness.

User avatar
gwaidan
Super Deluxe Wiggler
Posts: 1939
Joined: Sat Mar 07, 2009 7:26 pm
Location: Sinney, Orstraya
Contact:

Post by gwaidan » Thu Aug 23, 2012 8:05 pm

Carbon film resistors tend to have about -400ppm to -800ppm tempcos, whereas metal films are about +50 to +200ppm-the difference shouldn't matter except in oscillator cores.

I've found this before with 1K "metal film" resistors from Futurlec-in my case it was holding the resistor up to the globe in my desklamp, and finding out it had a negative temperature coefficient rather than a smaller positive one.

I ended up using all the "fake" resistors in projects that worked fine, but I haven't bought MF resistors from Futurlec again mainly because I like to know what I'm getting what I pay for. In their (and Tayda's) defence, they have the same vulnerability to this as any other dealer who buys parts from middlemen rather than with a traceable supplychain like Mouser, Farnell etc

Jarno
Super Deluxe Wiggler
Posts: 2841
Joined: Mon Nov 15, 2010 6:34 am
Location: Rosmalen, NL

Post by Jarno » Fri Aug 24, 2012 2:15 am

I checked out some datasheets for resistors yesterday, and in the KOA datasheet for CF resistors, for some values a negative temp coefficient was mentioned explicitly, in the same series, other values were specified as +/-
Also they defined a couple of resistance ranges and the exact value for temp coefficient was different between ranges.
But if we are talking room temp, say 17-26 degrees C that's really not that much variation, and it is likely to stabilise inside the modular synth enclosure.
The tolerance on the resistance value is a lot bigger than the tolerance due to the temp coefficient, so unless you select values, the circuit still needs to allow for those tolerances.

User avatar
L-1
Super Deluxe Wiggler
Posts: 2182
Joined: Wed Dec 14, 2011 8:20 am
Location: Belarus
Contact:

Post by L-1 » Fri Aug 24, 2012 8:32 am

I threw out all my fake mf resistors. Since I make a public projects, my builds must be reference for others. Fake have any datasheets and any parameters can't be guaranteed.
Multicomp 50ppm 100k drops to 98K
Fake 100K drops to 90-94K
Fake carbons with steel legs from my old stock drops to 80K
Before, I have built on the principle cheaper is better, now use only branded components.
When I heated fake ones their legs break off. There are such parameters as Load life and Terminal strength.

In general it's like wearing fake and say it not worst as armani. IMHO.

User avatar
L-1
Super Deluxe Wiggler
Posts: 2182
Joined: Wed Dec 14, 2011 8:20 am
Location: Belarus
Contact:

Post by L-1 » Fri Aug 24, 2012 9:42 am

I'm so adamant because my new projects are adapted and can be assembled for sale, and builders will make as I specified. So I specify - use branded. If Mouser or Farnell don't have some part, it should be checked. They don't have steel legs resistors mf or carbon.

User avatar
raisinbag
Super Deluxe Wiggler
Posts: 1708
Joined: Thu Aug 18, 2011 11:40 pm

Post by raisinbag » Fri Aug 24, 2012 9:42 am

Hey L1 I totally think you are approaching things the right way, I too use only the best quality raw materials in my products (precast concrete furniture) and (you have the coolest PCBs!!!!!) I wish the average consumer cared about who and what makes up the products they buy. It would be cool if electronics manufacturers listed all the individule componenets so that you knew if you were getting greatness or shit.

Well I originally wrote this post so I could sort through my confusion. I had followed some other threads on this forum where people were saying they had "fakes", I tried the "test" with soldering irons, lighters, etc to measure and compare know carbon and metal film resistors with the "unknown" "questionable" resistors I have. My results lead me to believe that I had "fake" metal film and I basically accused the place that I bought them from of selling me fakes and they responded assuring me that they are no fake. This is why wrote this post as I was so bloody confused by all of this.

So what it looks like I have in the end (assuming Tayda is telling the truth and I have to have a little faith in humanity) are metal film with a temp-co o the Whittier side of the scale.
This is one of those situations where I have learned a great deal, and thanks to the interweb and so many people willing to share information and experiences, we can sort through miss(diss)information. And hopefully by keeping a check on this kind of stuff we can all help each other from getting burned and ripped off.

Thanks all.

User avatar
raisinbag
Super Deluxe Wiggler
Posts: 1708
Joined: Thu Aug 18, 2011 11:40 pm

Post by raisinbag » Fri Aug 24, 2012 9:46 am

L-1 wrote:I'm so adamant because my new projects are adapted and can be assembled for sale, and builders will make as I specified. So I specify - use branded. If Mouser or Farnell don't have some part, it should be checked. They don't have steel legs resistors mf or carbon.
Ya totally man! When that Vocorder is ready (I have a DIY boner for it) I will certainly buy a speced BOM for this project!!!!!!!!! I will from now on buy branded resistors!!!! I have learned a lot from this experience!

User avatar
L-1
Super Deluxe Wiggler
Posts: 2182
Joined: Wed Dec 14, 2011 8:20 am
Location: Belarus
Contact:

Post by L-1 » Fri Aug 24, 2012 10:02 am

One more thing I noticed. IC sockets. Farnel write Contact Material: Beryllium Copper, Contact Plating: Gold. Not steel )) I think manufacturer don't just waste expensive metal.
http://ru.farnell.com/aries/08-3518-10/ ... dp/1674784

I think this applies to all fake parts. At the end everyone loves Buchla which I'm sure know what parts hi buy.

User avatar
andrewF
Super Deluxe Wiggler
Posts: 4061
Joined: Wed Apr 15, 2009 7:36 pm
Location: WA
Contact:

Post by andrewF » Fri Aug 24, 2012 11:33 am

:soapbox:

Probably it should be pointed out that for people who just wanna build some fucking mad synths and have fun making squelchy blip-blop noises at home and at a few gigs, it doesn't matter two fucks whether your resistors are MF, "fake-MF" or carbon. The circuits will still work fine.

Nor does it matter whether there is any gold or triple-dipped platinum fudgsticles on the chip sockets or the PCB traces. When you are simply trying to get something that sounds like robotic wet farts, a 2c socket will sound as good as a $20 socket, ears cannot tell the difference.

This is probably a weird idea, but a well designed circuit should be able to work perfectly well with crap quality components. L1 mentions the need to use the best components on his builds so his projects can be a benchmark for other builders. My view is L1 should build his projects with the worst & cheapest components he can find. If the projects work well with crap parts, then it is a kick-arse design. That is the real benchmark :razz:

lo-fi!!!!! FTW :nana:
/ :soapbox:

User avatar
cyklopljud
Wiggling with Experience
Posts: 366
Joined: Fri Mar 13, 2009 6:23 am
Location: Denmark

Post by cyklopljud » Fri Aug 24, 2012 11:36 am

andrewF wrote::soapbox:

Probably it should be pointed out that for people who just wanna build some fucking mad synths and have fun making squelchy blip-blop noises at home and at a few gigs, it doesn't matter two fucks whether your resistors are MF, "fake-MF" or carbon. The circuits will still work fine.

Nor does it matter whether there is any gold or triple-dipped platinum fudgsticles on the chip sockets or the PCB traces. When you are simply trying to get something that sounds like robotic wet farts, a 2c socket will sound as good as a $20 socket, ears cannot tell the difference.

lo-fi!!!!! FTW :nana:
/ :soapbox:
Well Said!

Post Reply

Return to “Music Tech DIY”