Whites Classic SL tuneup help?

[bklein]

A little help needed in getting a Classic SL up to snuff.
I have the IMD book and am referring to instructions on page 103. I also have a schematic located here: http://www.hobby-hour.com/electronic...tesclassic.php
The schematic doesn't exactly match my detector though. I have no VR6. I have VR3,1, and 2.
I see what appears to be 230 degrees offset of the TX/RX waveforms at idle.
The unit I have has some random or repetitive beeping that can be somewhat minimized by adjustment of VR3.
VR3 goes to R43 - but my R43 is 3.3K not 6.8K, so who knows how much of this schematic matches my board.
Paragraph 3 says to adjust Disc Offset until the sample pulse is located over the positive portion of the RX waveform.
Should I adjust worry about 30 degrees offset from stated? Should the sample pulse hit right at the peak of the RX waveform or where?
I have what looks to be ballpark signals but can't get rid of this motorboating/beeping. It does sense targets at respectable distances - penny at over 6".
It does detect its own coil cable. Discrimination works on pulltabs.
VR2 seems to go to U4-7. (?)
VR3 to R43. (bias for sample comparators)
VR1 to R46 and then to C22. (discriminator sample skew)
Perhaps if someone is familiar with this detector tuneup - just help out a bit.
Originally I thought I had a faulty coil because pressing on it made it sound. Took bottom cover off and don't see an issue - may have been shield ground intermittent.
I plan to get it working and add a coil connector for other coils I have for my XLT.
Some mods mentioned on this site too that I may try.

[Qiaozhi]

I'm not sure the Initial Setup Procedure described in ITMD will be any help in this instance. These instructions are for the Raptor Project.

If the White's Classic I is motorboating, then you'll need a scope to debug the problem. Have a look at the outputs (which are wired together) of the two comparators (U4A and U4B). I suspect you'll see some oscillations there. If so, then probe the input pins 3 and 5. From there you should be able to isolate which channel is the culprit, and trace back from there.

Also, try this schematic.

[bklein]

So by adjusting the two sample delay trimpots i can minimize the motorboating and still get detection and discrimination -but what are the test conditions and process to adjust these? I see the little diagrams and get something similar but how to tweak? Should there be no target in coil detect range or best to have one (what) near?
My TX signal and phase sample pulses are rock solid. RX signal has random and somewhat periodic modulation that I suppose could be EMI from my scope etc. on the bench - or something is faulty with the coil... I'll get a photo to post soon.
In a design like this what would the behavior be like if the sample pulses where much shorter? I don't quite see the reasoning for the long pulses.

[bklein]

Man, this is making me feel stupid. The TX is solid along with the sample control pulses but pin 14 of the LF444 has the variability causing the audio pulsing.
Pin 8 is relatively stable. The pinouts in my original schematic and the one above have some pin numbering issues. The original doesn't list U3D. The one above has input to U2B as pin one and it is really pin 4. Anyway, there are little switching glitches at the switching points that jump around with the variability of the preamp output signal. Is this where the amplitude variation is coming from? Is it a 4066 going flakey? I attach two scope shots - one with the input and another with the output side of this switch.

[bklein]

Here are two photos: one is U2B sample pulse pin 5 and U3A pin8, other is U2C sample pulse pin 13 and U3C-14.
Persistence shows the range of the trimpots. Does this look correct?
You can see the ripple on the preamp output.

[golfnut]

" but can't get rid of this motorboating/beeping"

A lot of these earlier whites machine are essentially the same.

The IDX I made does this sometimes, it can be quiet some days some days it grows and then sustains this motorboating. One of the manuals described this an a 'feature of the design, and normal'.


If you model the 2 amplifiers which are in series, (either the disc or geb ch, they are both the same ) the huge gain ~70dB and the phase response thru the 2 amps, they satisfy the basic criterion for a potentially unstable circuit. The 2nd amp output phase shift goes thru 180deg while there is still plenty of gain. So there is no phase margin.

Having an incredibly sensitive amplifier pair with a vast gain and no phase margin gives the mechanism for oscillation.

The amps are bandwidth limited high gain differentiators. The amp pair in a ace250 when modelled has phase margin and doesnt motorboat.


The freerunning motorboat frequency should be where the gain is maximum, about 8Hz

The GEB gate timing is preset on these, as is the Disc gate ( which is set to Disc out unwanted alloys out to just beyond aluminum Id suspect ) VR3 user control straddles the Disc and Geb channels, one way puts all metal thru on both comaparator inputs. The other way, the pot wiper selects the Disc output fully and user gets full preselected Disc. Or a mix of pot somewhere in the middle.

I introduced light resistive loads in the center of the amp pair which gave less trouble, I also used a transistor for the Diode pair ( biassed diode ) which helped.

Can be maddening, sometimes mine felt worse with a fresh battery. A good sprinkling of regulators would isolate stages and eliminate perfomance changes thru the battery range

[golfnut]

Forgot to mention, normally it is avoided to have 2 amps like this in high gain analog amps - due to risk of oscillation.

It is more common to have 3 amps in series, so the designer can ensure where the gain reduces to unity, the phase response hasnt reached 180deg so there is phase margin amd stability.


Some Integrated power amplifiers are either 1 or 3 stages for example. 2 stage amps can suffer instability and have critical layout, grounding, decoupling, screening requirements.

[bklein]

Thanks goldfnut, good to know that this can be common. From my photos you can see the preamp output ripple (red persistance). This variation seems to be what is causing the motorboating. It probably is at the 8hz frequency or so that you mention so does sound like this unit's problem. I'm thinking of trying another coil to see if it remains. The SL has a hardwired coil so I'd have to kludge one in or modify the box for the traditional connector.

[bklein]

The coil with the SL has some capacitors in parallel with the coils. I am suspecting it is the cause of my troubles but measuring it ends up with bogus data.
With my LC102 the tx is -(!)2.48uH and rx is 12.77mh. With a bk 875B the tx is 1.66mh and rx is 12.5mh.
Doesn't help much eh?

I've also come to realize that the spectrum xlt coil has a different connector than what I had intended to use (one like on gp extreme).
Anyone know the pinout?

[Silver Dollar]

Here is a typical White's coil schematic;

[bklein]

Thanks very much - never seen that detailed before.
What method is used to check such a coil? Frequency sweep?

Another problem I'm running into is obtaining the 5 pin xlr connector sets at a good price. Unique pinout for these Whites machines.

[bklein]

I wasted a lot of time working on this trying to find the cause of spurious comparator glitches when it turned out it was because my scope ground was connected to VCC/2 rather than Battery V-. I had been pulling IC's and socketing them to find the source.... ughh. I tried connecting my XLT coil and it didn't work well. Put the original back on and no motorboating! What fixed it - I don't know... Picks up a penny in air about 5". Since it doesn't work too well with the XLT coil I'll give up on that effort and just button it up and sell it.