Balanced to unbalanced board

[cspirou]

Thanks @luvdunhill! I'm using the buffer with the ubal/bal board which have a 1.3kohm resistor to ground on the output so i don't think an additional resistor is necessary. Plus I already have the PCBs and the pins match.

I am interested in buffers for another project though. Ill let you know.

[Pars]

cspirou: I would add an input resistor to ground as well. You can match Q1 to Q3 and Q5 to Q7, but I don’t think matching is all that useful here. If you want to try another small SMD buffer, let me know. I have just ordered some boards now that I have verified their performance versus sims.
 
I can include matched Toshiba JFETs and the other dual package BJTs are available at Mouser.
 

 

Those look pretty sweet Marc!


Sent from my iPhone using Tapatalk

[kevin gilmore]

what transistors are you using?

[luvdunhill]

what transistors are you using?

I am using 2SJ106/2SK208, BCM61/BCM62, PHPT610030-series.

The 2SJ106 is the SMT version of the 2SJ103. The 2SK208 is the SMT version of 2SK246.

[sbelyo]

On 1/20/2017 at 9:59 AM, kevin gilmore said:

it had better look like this

the output impedance is about 1k, working on the diamond buffer output now

I'm going to order the transistors for this and the buffers. 

1. I know that I have to match for hfe and Vbe but do I match all the N together and the P together or do I match N to P?

2. The highest hfe is the BC546\556 C, this only comes with formed leads.  Will I need to straighten them out, and if so I'm guessing there's no issue with that?

[jose]

What OPA is recommended? 

[sbelyo]

good question, it may say on the silkscreen.  I'll look at the board

Nope, I was hoping.  We'd have to ask kevin

[cspirou]

Should be the same OP27 used in the Dynalo

[jose]

In buffer and Un/bal board? 

[Pars]

The schematic I have says OP27.

diffinput3.PDF.pdf

[jose]

Thanks Pars. On our boards we haven´t any reference. I guess that we can use the OP27 on the buffer too. 

[sbelyo]

So should we match P to P and N to N?  Or that plus P to N?

[jose]

If nothing says otherwise, I will try to match as closed as possible P to P and N to N. 

[sbelyo]

That's what I was thinking

[sbelyo]

I ordered BC546B\556B  I couldn't find the C for both.  Anybody see a problem with the B?

[cspirou]

2 hours ago, sbelyo said:

I ordered BC546B\556B  I couldn't find the C for both.  Anybody see a problem with the B?

C has the highest hFe. B might be okay if you measure a bunch and take the highest measuring ones. It seems to be in stock here:

https://www.tme.eu/en/details/bc556c-dio/pnp-tht-transistors/diotec-semiconductor/bc556c/?utm_source=octopart.com&utm_medium=cpc&utm_campaign=compare-2018-01
https://www.rocelec.com/part/FAIFSCBC556CBU

I haven't ordered from either place so I don't know what they are like.

[sbelyo]

That first place had them both...  Darn, guess I didn't search hard enough.  I ordered 250 each of the B from arrow and a transistor tester off ebay.  We'll see what kind of numbers I get.  

[sbelyo]

the transistor tester measures hFE and Vf.  What's are acceptable values for this circuit?  How closely should I match the devices? 

 

[Pars]

I’d match for hFE, but also try to get Vf (also known as Vbe) as close as possible as well. I’d shoot for 5% if possible. Matching the N and P is often nigh impossible depending on device type. You just have to do the best you can.


Sent from my iPhone using Tapatalk

[sbelyo]

Got it...  Is an hFE of 300 good enough for this circuit?

[sbelyo]

The majority of what I'm seeing is 315 hFE after measuring 34 devices.  Is 315 good enough?

[Pars]

I would think so. That is in the upper half at least of the range for B devices.

[sbelyo]

cool...  I should be able to get them all tightly matched then.

[jose]

Finally, can BTAs work?

 

[Pars]

BTA is the newer B designation.

Just wondering if MPSA06/56 could be used in place of these? Pinout is backwards (EBC vs. CBE for BCs).