| Author |
Message |
reve

Joined: Feb 23, 2008 Posts: 149 Location: USA
|
Posted: Fri Nov 18, 2011 12:12 am Post subject:
minimoog vcf: c4-c7 values & transistor hFE Subject description: i'm thinking about using 3904s. |
 |
|
Hi Yves et al!
So after a number of successes building modules of Yves', I picked up a pair of Moog filter PCBs from the Bridechamber.
Being American, I have a lot more 3904s on hand than I have BC547s. I figured I'd just match those, use sockets (reversing them, of course) and if it's not optimal I could just pull 'em.
But after trolling old forum posts, I saw Yves said he generally used the (higher hFE) BC547C in non-VCO designs. But as I understand it, the optimal value of c4-c7 was determined by the hFE of the transistors used in the ladder. Presumably my trannies will have a lower forward current gain, but I did note that Yves didn't specify a suffix for the transistors in the parts list.
So: does this matter? Am I being pedantic? Or should I just match some 3904s, solder in 47n caps and not worry about it?
Thanks for reading! _________________ - r. mosquito
SynthDIY Synthpop -> http://www.reverbnation.com/photovoltaik |
|
|
Back to top
|
|
 |
iopop3
Joined: May 28, 2010 Posts: 94 Location: Malmö, Sweden
|
Posted: Fri Nov 18, 2011 12:29 am Post subject:
|
 |
|
I built one of these recently. Put off the matching process for so long that I said, screw it, and just used what I had at home. BC547Bs, which I presume are for the same batch, got them loose in a bag from my supplier. Did match the caps though.
Everything seems to work ok, I do get it to self oscillate and it feels pretty stable when checking via scope.
What impact does the matched transistors have, temperature stability? |
|
|
Back to top
|
|
 |
reve

Joined: Feb 23, 2008 Posts: 149 Location: USA
|
Posted: Sat Nov 19, 2011 12:31 am Post subject:
|
 |
|
| iopop3 wrote: | | What impact does the matched transistors have, temperature stability? |
I'm not an EE, but I reckon that by matching the transistors (and caps) you get identical currents on both sides of the ladder... Different currents means messing up where the filter poles appear on the s-plane? Which would mean a different transfer function and a different sound?
Dunno, but I'm pretty sure it's not temperature stability. _________________ - r. mosquito
SynthDIY Synthpop -> http://www.reverbnation.com/photovoltaik |
|
|
Back to top
|
|
 |
varice

Joined: Dec 29, 2004 Posts: 961 Location: Northeastern shore of Toledo Bend
Audio files: 29
G2 patch files: 54
|
Posted: Sat Nov 19, 2011 3:00 am Post subject:
|
 |
|
| iopop3 wrote: | | ...What impact does the matched transistors have, temperature stability? |
I’ll admit that I’m not an expert on the Moog ladder filter, but from what I gather it appears that matching the transistor pairs for Vbe in the ladder helps balance the current on each side of the ladder which results in reducing the amount of filter cutoff CV bleed through at the output at the top of the ladder.
On the other hand, a close match of the two transistors in the ladder current source (as Q1 and U1/Q5 in the YuSynth Minimoog filter clone) helps with temperature compensation to stabilize the initial cutoff frequency of the filter.
| iopop3 wrote: | | ...Did match the caps though. |
Are there specifications for matched ladder caps?
I haven’t seen anything yet, but I would be interested to find out if close Vbe matching of all the ladder transistors along with closely matched ladder caps would make the filter sound the best. Maybe not though, as some “loose” designs still seem to sound good. _________________ varice |
|
|
Back to top
|
|
 |
iopop3
Joined: May 28, 2010 Posts: 94 Location: Malmö, Sweden
|
Posted: Sat Nov 19, 2011 3:44 am Post subject:
|
 |
|
Thanks for your answers, it does sound logical. I was probably thinking too much about VCOs when it comes to matched transistors and temperature stability.
| varice wrote: | | Are there specifications for matched ladder caps? |
Yves list that the 47nF should be matched to 1%. Haven't seen any other documents or info. |
|
|
Back to top
|
|
 |
reve

Joined: Feb 23, 2008 Posts: 149 Location: USA
|
|
|
Back to top
|
|
 |
yusynth

Joined: Nov 24, 2005 Posts: 1314 Location: France
|
Posted: Sun Nov 20, 2011 5:25 am Post subject:
Re: minimoog vcf: c4-c7 values & transistor hFE Subject description: i'm thinking about using 3904s. |
 |
|
| reve wrote: | | Being American, I have a lot more 3904s on hand than I have BC547s. I figured I'd just match those, use sockets (reversing them, of course) and if it's not optimal I could just pull 'em. |
There is no problem using 2N3904 as substitutes for the BC547 (just keep in mind that the pin out is reversed when you solder them of the yusynth PCB). Here the suffix C has little impact except that it offers a wider current range and as such facilitates the obtention of a wide frequency span and also because this is the brand I tested and that I can recommand.
Concerning the matching it is important to match the transistor pairs and also the caps (do not buy expensive caps for this but just select them in a batch of caps) if you want to have an extended range of auto-oscillation. Not matching will reduce the auto-oscillation range. By the way matching the absolute value of the caps is not really important, what is really meaningful is to have the values of the 4 caps within a 2% range. _________________ Yves |
|
|
Back to top
|
|
 |
|