ROG Flipster works great! (but better with proper bias?)

[zjokka]

In order to find an introductory project for ulitmately being able to build a decent Rhodes preamp, I couldn't resist building the Flipster. Especially after so many raving reviews.

So built it immediately on perfboard, using J201s and one 2N5457 (Q4), I gathered from the report that this would yield better results. Was the most complex build for me yet, but up until now everything worked immediately (except the beginner project ). So was sure or hoped  I would mess up to get some debugging exercise, because I had trouble biasing some working circuits ...

So was glad to measure or set 4.5V to the first 3 trimpots, but the last one jumped from 3V to 8.8V with a microscopic twist. Had even trouble putting it back on 3V at a certain point.Nelson solved this with a larger (500k) trimpot, which I don't have for the moment. .

But I did test the circuit, and it's runs just fine at 3V. I do seem to get a little double sound, but this might just be the distortion, or: the bass. I had this Cort active bass lying around, I though I could 'passivise' it by just ripping out the battery, but elas! I put it at the lowest preamp volume, so don't really know maybe the signal is too low. Will get another bass in the picture, but hey, it's sunday.

I tried biasing everything at 6V as is also sugested on ROG, but then Q4 was too far off and volume very low. I did forget the the polarized 100uF cap (9V to ground) that on the schematic but not on the  layout. Would that influence matters?

tungngruv even has different bias voltages although we uses similar JFETs throughout

Would bet better to bias all points at the same 3V as would be possible now, or do I lose a lot of EQ flexibilty with that? I'm really curious as to how biasing the JFETs really affects the circuit.  Also, I find it very hard to tell what is acceptable noise level, but will also have to try the 0.1uF input cap that was also suggested.

My ears say it rocks, but my mind say I might do (bias) better.
What do I do? 

ZJ

[Bernardduur]

I had the same problem with the bias; I just continued until I got the right value..... even breathing at the trimpot did change the value

It sounds better

[zjokka]

thanks,

will give it another go later tonight. but if i understand the matter correctly, it's not a result of the trimpot. Or should I take out the trimpot  en substitute for another? I have other 100ks lying around.

ZJ

ps: thanks for suggesting the Kreuzer preamp, will keep it in mind; too complex for me at the moment.

[Bernardduur]

I dunno; I never thought it was due to the pot..... maybe it was. I just got the desired value by a lot of tweaking, a lot of turning and a bit of luck.

Great pedal btw!! I use it almost all the time to add some grit to my bass tone

[zjokka]

Do you have the bias at 4.5V as suggested intially, or higher?
Do you know how higher/lower biasing affects the circuit?

sorry, so many questions
thanks still!
ZJ

[Bernardduur]

I don't know WHY it is 4.5V, but I know I LOVE the sound on 4.5V and disliked it at other settings.......

I've also tried it at 6V (I guess) but I more liked the 4.5V; the grit is more real

[Ge_Whiz]

In general, a 2N5457 is difficult to bias with a 100k trimmer; it needs a much lower resistance, often only a few kohms. Substitute a 22k or even a 10k trimmer to bias it.

[zjokka]

In general, a 2N5457 is difficult to bias with a 100k trimmer; it needs a much lower resistance, often only a few kohms. Substitute a 22k or even a 10k trimmer to bias it.

That's a lesson learned for the future. Had no other JFETs. Still some desoldering to do.

How about that dubious capacitator at the power source?

I did forget the the polarized 100uF cap (9V to ground) that on the schematic but not on the  layout. Would that influence matters?

is it necessary?

Would bet better to bias all points at the same 3V as would be possible now, or do I lose a lot of EQ flexibilty with that? I'm really curious as to how biasing the JFETs really affects the circuit.  Also, I find it very hard to tell what is acceptable noise level, but will also have to try the 0.1uF input cap that was also suggested.

would there be any reason NOT to put the input cap? I don't hear a terrible amount of noise....yet/in my test setup/at headphone levels?

Will report back soon with the results of the lower trimpot. Sure hope 10K is enough, don't want to desolder twice. heck, might measure and put a fixed resistor its place.

thanks all so far
ZJ

[analog kid]

Have this problem often w/ my JFET amp schem fx.  I'll have maybe one or more fets whose bias vltg will jump from very low to like 7v +  in that one minute little twist. I first experienced and drove myself nuts for a few minutes thinkin it might be my cheap little metal trimmers , until went OH lets try some different FETs in these spots and bam! Smooth even trimming. So It may depend on the circuit I'm sure , but fets are very unpredictable like that I think. Of all types of trannies I think they are ones that it pays to have a big bag of whatever you are using.        I am lucky enough to have gotten samples of 5457, 5458, J201 , MPF102 and a couple others a long time ago so I just swap freely from my bags.     
So long as you can get it to that target bias vltge no matter what a pain to tweak it , I'd say it doesn't matter. I didn't notice a negative difference in the fets that were hard to get trim to land on 4.5 against the ones the the trimmed smoothly over and under 4.5v.  So get it there and leave it.   But There are some character differences in higher /lower bias for these circuits though the heavier gain one's especailly I think.  Use fixed resistors and toggle to have access to a couple different bias vltges.    thinking of that, Look at the Variac mod on the thunderchief

[zjokka]

thanks kid!

followed the earlier suggestion (my electroshop doesn't anything but the j201s) and swapped out the 100k trimmer for a 25k and the 2N5457 biassed easily.

Q1 and Q4 are right on target now, but I notice that Q2 and Q3 have opposite running voltages. I mean, if Q2 measures 4.2V and is slowly rising by 2mv/sec or something, Q3 will be decreasing at the same speed from say 6.5V.

Will bench check the circuit plugged in tonight. Even before I could bias Q4 2n5457 with the new trimpot, it worked allright.
Last few night I biased, left it a night, came back to find different values for Q2 and Q3 and on the move.

Is this normal? Is there some capacitance going on or something? These value should be steady, no?

ZJ

[Bernardduur]

Well, I have the issue that my values change when I bias a pot, but when I work from left to right I don't have those issues.

[zjokka]

thanks Bernard,

I tried working from any direction, the result doesn't change: outer trimpots stable, inner unstable. When I had the 100k trimpot on the 2N5457 the middle two had no problem at all. At least now I got three of them stable and one hovering about the 4,5V I benchtested throu a mixing board, and seem to do fine.

Will be boxing this one up soon I think. Not going to make it a regular stompbox but rather a unit that would sit on top of the amp or electric piano with a on/off and boost/true bypass switch.

thanks still,
zj

[Xavier]

I'm bumping this thread because I'm having the same problem with the Flipster I'm building for my bass player. In this case is all 2N5457 in all 4 positions. All four 100K trimmers

Q1: No problem, it biases correctly
Q2: 9 volt throughout , doesn't change (bad trimpot?)
Q3: Goes from 1V to 9V just by looking at it
Q5: No problem.

I'm thinking about using fixed resistors.

And I haven't even heard it...............boy I'm frustrated. Even my Obsidian-T hasn't worked properly. I'm about to give up.....

[Bernardduur]

Use smaller valued trimpots; worked like a charm for me

[Xavier]

Use smaller valued trimpots; worked like a charm for me

I'll try with 22k trimpots and fixed resistors, thanks...................... I really want to hear this one

[petemoore]

So was glad to measure or set 4.5V to the first 3 trimpots, but the last one jumped from 3V to 8.8V with a microscopic twist. Had even trouble putting it back on 3V at a certain point.
  Yupp...often is the case.
  I take a resistor or two...say between 15k or 47k, strap that from V+ top rail, touching the Drain [which is across the 100k pot, reducing it's value, making the R sweep cover a smaller range from CW to CCW].
   Then when I get nice adjustment, I trim, and stuff the resistor lead into the extra socket lug [I split 8pin IC Sockets, using the bottom three lugs for S/G/D, the top lug for that extra drain resistor, when soldering the drain lug, I just include the 4th top lug of the socket for that purpose. It is possible to finnesse a resistor lead and transistor lead into the same socket lug.
  Swapping Jfet is probably easier.
  I have had bad trimpots, they are a bit delicate I guess, but easily measured with the DMM to affirm function/nonfunction. Sometimes adding a fixed resistor 'moves' a pots dead spot to where you 'want it'...so the good portion of the pot is setting/adjusting to proper bias...IOW test and adapt to funky trimpots when necessary.

[Xavier]

And wouldn't it be easier to use a voltage divider with, say , 2x 1M resistors and hit each drain with the friggin' 4,5V already? I know it sounds stupid, but wouldn't this save the need (and physical space) of the four trimmers?

[Xavier]

Is it normal that all transistors bias at around 10k-20k but Q2 biases at 10M? (4,5V in all drains)