little Help troubleshooting diy Echoflanger

[analog kid]

Hey everyone,
I was really hoping that if I just broke down and posted concerning this little issue I've been wrestling with on my recently completed Echoflanger project , that it just might get the attention and bring out some assistance from some of you guys out there who I KNOW are familiar with these ol' circuits ( you know who you are ; )  just kidding  no help is obviously welcome from ANYONE who has an idea at all of what the issue is!
   OK , first I hope it's alright to link to a place where I know the schem is at ( if not someone tell me and i'll remove it ) since I don't know how many folks would have the EF 2page schem convenient.    Munky's site btw (THANKS)

http://www.geocities.com/diygescorp/schematics.html
  Ok I built this beast from this schematic and from layout of the actual original unit which S.Giles( I believe) scanned , and Paul Nelson very nicely cleaned up etc...  as well as referencing to pics of actual unit to look for any mistakes in parts values etc... there were a couple I am pretty sure. But for TSing purposes , after my Feedback issue showed up I did pretty much try all those parts values that are listed with different values.

Short Report~~
Ok , All modes are functioning as they should be Flange/Filtermatrix gives a very nice sweep when the clock is set fast enough , very wide actually (I think  I got around 100kHz-1.2mHz...suprising!)   and the Slapback , and chorus  CAN BE very strong IF of course , the clock is set low enough.  So there is a compromise that one must make when setting up the unit.   Sacrifice from the widest swept flange you can get to get a stronger slapback repeat VS.  set flange for optimum sweep and deal with a Slapback that is a bit weak.   Surprisingly also , the Chorus still sounds great even when compromising the lower clock freq.

HERE'S MY PROBLEM~~
I am experiencing a Feedback issue that is a tad hard to explain but I'll try.  It's most offending in Flange/F.Mtrx,  and is affected by Input signal  and frequency selective. Also maybe w/o saying but only existing when FB is turned up regardless of freq,etc... Pretty much I have a feedback better described as a strong piercing Tone  which only begins to become "audible as this"  only when sweep hits below apprx 200kHz.  When frequency reached above 200khz Piercing hi freq  smoothly turns into normal sounding "regeneration" within the signal RATHER THAN On TOP of It.    And here's the strange thing to me,  It's sort of  reverse "gated" by the input signal.  So it's only wailing  when  input falls or Playing stops .  (mimics high volume speaker fb very well    too bad it's not desired)  Playing steady and consistent will keep "That FEEDBACK" away completely out of the audio as long as you want it to but when input signal falls it's hideous      In SlapB and Chorus this is much less a problem although the action of the input level and freq are the same , the FB isn't audible as the same piercing Hi Freq TONE at any width/tune settings which I find a bit odd because these modes should be at even lower freq's than the Flange is  when it's exhibiting the "FB tone".   

Now as I said it's frequency selective, so  in normal LFO flange mode this is heard only as a brief "Whistle" as the low freq's are cycled through , for how long depending on Rate.   BUT in Manual Flange (Filter Matrix) , which Is how I've been troubleshooting it cause you can zero in on the offending FB,  it's an even bigger problem because when I sweep it into the lower freq range with the TUNE and WIDTH controls  , this leaves the Feedback /Oscillation  on steady.  Again , as long as input signal is not present (or very low I think)   Striking a chord / note Kills the high freq feedback until note decays again.

I've tried several things that I could think of as well as friends whose opinions I respect and to no avail so far.  Such as different dual op's for IC1/ 2  and different NE570's and several cap value tweaks but basically I'm hoping that if someone doesn't just know from the get go what this problem is, and I'm pretty much assuming it can't be normal with the Echoflanger,   that at least maybe I can be led in the right direction as far as caps to look at or where to probe
I have checked multiple times my components/values but you never know.  so If it would still happen to be a bad component or value at least someone can maybe get me looking in the right area of the crkt where this symptom like this would be caused! I pretty much don't have that narrowed down now at this point     I REALLY APPRECIATE Any Help I get here
  Thanks ALL!!

[analog kid]

BUMP...
    and btw I should point out that even within the frequencies where the Offending feedback "tone" is produced,  I can trim the feedback trimmer and /or series trimmer with NE570 compressor  to the point where the fb pitch is  no longer offending but of course by then the overall feedback at any settings becomes so weak that it's hardly there.  SO that's not much of a fix.    At least I have a seemingly perfect working EF as long as no feedback is engaged.   

and since slapback and chorus are much less affected by this feedback issue  , I can pretty much engage as much FB as I want. and manual flanging so long as I don't set to a frequency below about 180-200khz.
  It's just pretty obvious there's a problem somewhere, I would just love some advice from someone acquainted with this crkt as to WHAT SECTION of the crkt a problem would likely cause a fb symptom like this.              OR ALSO from any EF owners , On how their units behave in relation to this problem.

THANKS AGAIN