[sdiy] Transistor matching Was: Another satisfied Takuya Katayama Tempco customer.

Sat Jul 19 18:17:44 CEST 2008

On Jul 19, 2008, at 5:44 PM, cheater cheater wrote:

> Hmm, I assume just testing each transistor on its own and recording
> parameters wouldn't be good enough.
No
> You really need a circuit of two
> transistors, don't you?

Yes you pick two random transistors and compare the two with your dual  
channel oscilloscope. Superimpose the curve and give them a note let's  
say from 1 to 10 . Then you would bag them and take the best matched  
ones for your projects.
I even have a few extra pre matched bags with labels saying 8.5 or 9  
or something on them . That way next time I don't even have to match  
them. It's already done. But it's really fun to do it and easy as well.

>
>
> How would you go about testing each pair? You'd have to have some sort
> of serving facility to store and serve the parts somehow.
>
> Best regards,
> Damian
>
>
> On Sat, Jul 19, 2008 at 4:19 PM, Csaba Zvekan <czvekan at gmail.com>  
> wrote:
>> Damian,
>>
>> Thanks Damien, ..lol...Yeah I didn't think of that. Labeling &  
>> packaging of
>> the super matched  transistors . :)
>> You will laugh cause I really have some barcode readers laying  
>> around for a
>> couple of years now. I remember scavenging an old CD label / sorter  
>> factory.
>> That's where  I got them from  .Up until now I had no idea what to  
>> do with
>> them. :)
>> No, but seriously that transistor project  is awesome and credit  
>> goes for
>> that wonderful man who designed the thing.
>> Also I am nothing compared to these people on the SDIY list. They  
>> are the
>> real miracle makers . I am just a good copy cat :) and feel like  
>> information
>> and knowledge should be free and shared. Cause at one point it was  
>> freely
>> shared to me.
>>
>> Csaba
>>
>> On Jul 19, 2008, at 3:11 PM, cheater cheater wrote:
>>
>>> Csaba,
>>> I assume it would also have to label each part that's being  
>>> matched to
>>> keep an inventory. Some sort of barcode system, presumably. So,  
>>> you'd
>>> need a barcode reader in there as well.
>>> All your projects seem pretty cool :) I'm quite impressed. And if  
>>> you
>>> do make an automated matching facility, make sure to let us know :D
>>>
>>> Cheers
>>> Damian
>>>
>>> On 7/18/08, Csaba Zvekan <czvekan at gmail.com> wrote:
>>>>
>>>> Ohh that's no problem I also thought of that as well . :) ...lol...
>>>>
>>>> see my DIY CNC/ SMD pick and place project :
>>>>
>>>> http://www.youtube.com/watch?v=hOVnKK6m5a4
>>>>
>>>> ...lol..
>>>>
>>>>
>>>> Just kidding it just mills and drills ... could be modified   
>>>> though :)
>>>>
>>>> Csaba
>>>>
>>>>
>>>>
>>>> On Jul 18, 2008, at 10:00 PM, cheater cheater wrote:
>>>>
>>>>
>>>>> The only thing we need now is to come up with a robotic arm to  
>>>>> load
>>>>> transistors from SMD mounting tape, then we can run&forget for two
>>>>> weeks and have a very nicely matched bulk of transistors. :)
>>>>> I assume this is what companies normally use to match them?
>>>>>
>>>>> Cheers
>>>>> Damian
>>>>>
>>>>> On 7/18/08, Csaba Zvekan <czvekan at gmail.com> wrote:
>>>>>
>>>>>> Ian ,
>>>>>>
>>>>>> I am by now means an electronic guru . And can not answer your
>>>>>> question.
>>>>>> All I know is that I have matched some transistors (BC547) for  
>>>>>> a VCO
>>>>
>>>> project
>>>>>>
>>>>>> (Yves VCO Yusynths
>>>>>> http://yusynth.net/Modular/EN/VCO/index.html
>>>>>> and compared with the LM394 SuperMatched Pair) . And find no  
>>>>>> noticeable
>>>>>> difference in temperature drift or anything.
>>>>>> The only difference is the price  compare for yourself LM394 is  
>>>>>> 22.35$
>>>>
>>>> and
>>>>>>
>>>>>> two BC547 is 0.20$ . Ok you would still need to purchase a  
>>>>>> bunch to
>>>>
>>>> compared
>>>>>>
>>>>>> them. But you get the idea.
>>>>>> All though I must say that the precision OP-AMP OP490  is rather
>>>>
>>>> expensive
>>>>>>
>>>>>> ( 15.70$ @ farnell.com )  :( but hey , we want precision,  
>>>>>> right ? And
>>>>
>>>> it's
>>>>>>
>>>>>> not like we need to buy hundred of them. The rest of the  
>>>>>> project is
>>>>
>>>> rather
>>>>>>
>>>>>> cheap .
>>>>>> Make sure you handle measuring in the morning and touch the  
>>>>>> transistors
>>>>>> with tweezers. Also give it some minutes to settle . You 'll  
>>>>>> get the
>>>>
>>>> hang of
>>>>>>
>>>>>> it.
>>>>>> The original design I got from here:
>>>>>>
>>>>>>
>>>> http://electronicdesign.com/Articles/Index.cfm?AD=1&AD=1&ArticleID=6251
>>>>>>
>>>>>> Csaba
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Jul 18, 2008, at 4:24 PM, Ian Fritz wrote:
>>>>>>
>>>>>>
>>>>>>
>>>>>>> At 05:29 AM 7/18/2008, Csaba Zvekan wrote:
>>>>>>>
>>>>>>>
>>>>>>>> We SDIY people always need matched transistor pairs :
>>>>>>>>
>>>>>>>> that's why I recommend this:
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>
>>>>
>>>> http://www.csaba.ch/index.php?option=com_content&task=view&id=14&Itemid=30
>>>>>>
>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>> Is there some advantage of this over a simple bridge circuit?   
>>>>>>> How
>>>>>>>
>>>>>> precisely can you measure the matching?
>>>>>>
>>>>>>>
>>>>>>> Ian
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> Synth-diy mailing list
>>>>>> Synth-diy at dropmix.xs4all.nl
>>>>>> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>>>>>>
>>>>>>
>>>>>
>>>>
>>>> _______________________________________________
>>>> Synth-diy mailing list
>>>> Synth-diy at dropmix.xs4all.nl
>>>> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>>>>
>>> _______________________________________________
>>> Synth-diy mailing list
>>> Synth-diy at dropmix.xs4all.nl
>>> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>>
>> _______________________________________________
>> Synth-diy mailing list
>> Synth-diy at dropmix.xs4all.nl
>> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy