Farfisa Compact Mini Divider problems

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

tardishead

Well-known member
Joined
Aug 11, 2004
Messages
623
Hi there
I'm restoring a mini compact Farfisa
The schematic for the divider boards is here
http://www.bustedgear.com/repair_Farfisa_minicompact_dividers_4.html
After replacing loads of Ge transistors and finding a hairline crack in one of the cards across all the pcb traces - took hours ouch!!!  I got her up and gave it a tune
Sounds awesome
But then after using for about half an hour two of the cards at one point in their divider circuits gave a tone a fifth lower than it should be.
There are not many parts in these divider sections. I checked the transistors and polystyrene caps - ok. Some of the resistors seem to be quite out of tolerance. But can someone explain to me what can be causing this. In the divider sections - the resistors on the collector of the first transistor and the base of the next one - do they set the increment of division? And if these are out of tolerance could that result in this extra interval of division? Also does old resistor tolerance get worse with heat or useage?
:-\
 
I restored a Farfisa last year. Fun project and they sound awesome once healthy.
A fifth is strange. An octave off is a more common problem in these type of circuits - with a bad divider.
As I understand it, the initial tone is set with a few components - a resistor (A to B in the schematic) and the variable inductor. Then the pitch is divided down by octaves in the 5 divider sections. A bad transistor can cause a faulty divider.
If it is a fifth off I would look at the resistor that sets the appx range and then the variable inductor. Hopefully the variable inductor isn't the problem as those are a custom part. I had to replace one and got it off ebay.
Have you changed / tested the electrolytics?
 
No fixed all the bugs and then this happened at 2 different cards 3 octaves down in the divider chain
I've debugged tonnes already. Inductors all work well
Can someone point me to the components in the divider sections that set the divide frequency? And preferably explain to me whats going on
 
The top octave is set with the 12 inductors as part of the oscillator circuit, so if the top 12 notes are playing right, those aren't the problem.

I remember the circuits because I had a MiniCompact as a teen. Each stage (two-transistor divider circuit) is basically a relaxation oscillator (rather than a flip-flop, which always works with any input frequency), and only oscillates over a certain range of frequencies (it'll oscillate by itself with no input, and only sync to an input over maybe half an octave at most). The associated resistors and capacitors determine the frequency. Notice that the capacitor values approximately double for each octave down, and they are (probably, but I don't remember for sure) different for the same positions in every divider board. I recall the resistors are like 10 percent tolerance, and the capacitors would be in that range as well. Any changes in these components can make the divider marginal and "divide" by something like a 3-to-2 ratio rather than 2 to 1. You could try paralleling some other resistor values across the relevant resistors, or capacitors across the existing capacitors. Not sure how to tell you what the proper place or values are, but if it makes it work, you've 'fixed' it.

I suspect a leaky transistor or one with low gain could do this as well - I recall germaniums have a bad reputation for high leakage current.
 
Fantastic Ben
That's great info for me to work with.
Is HFE important in these relaxation circuits? Should pairs of transistors be matched for HFE? I'd there a minimum HFE that will work?
 
From my memories of fixing a similar organ years ago, this fault was always caused by faulty divider transistors. I think the manufacturer bought in enormous bulk but poor quality! Ge transistors are notoriously bad for leakage. If I were fixing this I would swap transistors first and only look at Cs and Rs if the fault persisted.
 
Ok so I replaced some transistors with 2n3906. On one card where I replaced at least half the transistors the tone did change and become a bit lifeless compared to the other notes.
So I figure I have to look into some germanium replacements. Which seems to be a minefield!  I took one original out checked HFE which was 250. Pretty high I thought but then realised that's not factoring the leakage current! So HFE could realistically be half that.
How important is HFE in these circuits?
If it's not  a problem to have low HFE then there seems to be quite good deals to be had in Russian germaniums. But anything with tested high HFE seems to be expensive.
AC125,oc71 and sft352 seem to be the parts required
Can someone help?
 
OC71 is hFE rated >40, 13uA c-b leakage. I would not want high hFE. I would not worry much about leakage. Talk with Steve at SmallBear. He buys bulk lots of Ge devices and tests-out to find works-like XYZ parts. If you want minimum-spec "OC71" he can probably round up dozens at far less than the prices asked for "genuine OC71".
 

Latest posts

Back
Top