MFJ-989D Tuner

turns counter adjustment, meter board shielding…

© 2018 by KV5R. All Rights Reserved. Rev. 2/9/2019.

Introduction

The MFJ-989D Versa Tuner V is MFJ’s largest manual tuner. It is rated at full US legal power, and features a roller inductor, active peak-reading watt-meters with LED lighting, antenna switch, dummy load, and a big 1:1 current balun. So it puts several station accessories in one big box.

MFJ is well-known for making ham radio accessories that are of reasonably good quality and price. They are also known for occasional loose parts and sundry other little mechanical problems. Regular purchasers expect to give things like antenna tuners a good internal inspection and, if needed, adjust and tighten a few things.

Turns Counter Slippage Repair

My new 989D tuner worked fine right out of the box, except for occasional slipping of the turns counter gears. After taking hours to build a tuning chart, I discovered the counter was off, even though I had zeroed it before starting. Bummer.

A while later, the meter circuit started misbehaving, and it took me two tries to get that one fixed. A while after that, I had to replace the LM324N quad op-amp IC on the meter board.


MFJ-989D Versa-Tuner V.


Roll it fully counter-clockwise (don’t bang the stop) and check the counter. Pushing a paperclip through the little hole will zero it. Unfortunately on mine (and many others, according to the reviews) the plastic gears will slip when spinning the roller inductor, so the counter needed to be zeroed again. Unacceptable! If a counter isn’t reliably repeatable, it’s useless.


The plastic counter is mounted on this L bracket, along with the 9V battery holder.


The bracket was not bent a full 90 degrees, so the counter gear shaft is misaligned and the gears are not fully meshed. They are barely even touching! Can you imagine aligning and tightening those gears and not noticing that?


Pressing the bracket shows where it should be.


The big gear will need to be realigned to the new counter position. Loosen both set-screws using a long 1/16th-inch Allen wrench.


Okay. Take it out.


Remove counter from bracket. Note that one hole is slotted for adjustment. Note also that one screw is missing, as there’s no scuffing on the bracket.


Out to the workshop with it! The framing square shows the problem.


Put it in the vise and give it a couple whacks with drift pin and hammer.


That’s better!


Now let’s put just a bit of grease on the worm gear…


…and the big gear. Caution: some types of grease will melt nylon, so look it up and use the right kind. Silicone and lithium grease should be safe to use. Do NOT use any type petroleum grease, or a synthetic diester grease! Actually, plastic gears do not need lubrication, but I used it just to make them a little quieter and smoother. It did.


Reinstall the counter assembly. Do not tighten the counter-to-bracket mounting screws until the gears are properly meshed. Do not press the gears tightly together or tuning will be rough and bumpy. Just bring the gears together, fully meshed, and tighten the screws. Do not over-tighten! This is fine-thread screws into nylon.


Align the big gear centered on the small one and snug its set-screws. Roll the inductor around and test the feel. Adjust the counter position again if necessary. Wipe off any excess grease that squeezes out of the gears.


While testing the gear meshing, I noticed that the shaft wobbles a little too much. This flexing puts stress on the counter gears on the up-swing of the crank, and spreads the gears a bit on the down-swing. I slipped in a bit of heat-shrink tubing (not shrunk, it’s tight enough) and it took up most of the slop in the shaft bushing.

Extra Views

While it was open, I inspected the internals and make a photo record.


Here’s the dummy load resistor. It’s a 50 Ohm, 25 watt (continuous) non-inductive resistor, 300 watts max at short duration (30 seconds max, 2 minute cool-down). It seems rather silly to put a 300 watt dummy load in a 1500 watt tuner. The manual warns, never transmit into it with an amplifier! Buy an overpriced 1500 watt dummy load, or build a salt-water dummy load (see my Salt Water Dummy Load article).


Here’s the balun. I was very pleased to see it’s the stacked-core, bifilar-wound 1:1 current balun/choke that’s now recommended by various balun experts for ladder line. Yay. Perhaps the days of misusing 4:1 baluns (and the burning up thereof) for ladder line are finally over.


The Air-Core Roller Inductor. Frankly, I’d rather have a good ol’ ceramic-core, but I guess they’re no longer available…


One of the two giant capacitors. After putting it back together, looking at the pictures I noticed that this one needs a bit of adjustment, as the plates are not perfectly centered. And I did, after the next repair.


The Bruene directional coupler for the meters.


The metering board. Several people in the ham forums complained about the stupid placement of the board such that one of the calibration adjustment pots is right under the antenna switch shaft. Apparently they later put holes in the bottom of the case, and the board, so they may accessed from below.


The antenna switch uses two decks in parallel. Looks okay. The leads could be shorter.


A common problem with MFJ cabinets. One would think that after many thousands of such cabinets, they might adjust their metal-working machinery a bit, so that screw holes align and counter brackets are properly bent…

So how does it work?

It does what an antenna tuner should, and after correcting the counter bracket, is very repeatable when using a tuner chart. It has not slipped one bit since I fixed the bracket. I made a table of settings every 25kc across all the bands, and on my 80-meter doublet it tuned everything flat except a portion of 28 MHz, which I never use. A little change in feed line length will usually move a problem spot out of the band.

Later: RF Gets in the Metering Board

This problem didn’t show up at first, because with my 130-foot dipole the transmitter-side capacitor was at around 4 on the dial, and the capacitor plates were far away from the meter board. But after I cut my antenna down to 123 feet, the cap setting was 7-8, and its plates were very close to the board. What happens is that RF gets into the metering circuitry and causes an op-amp on the reflected side to gate full-on. The reflected meter pegs out, even though the power is only about 250 watts and the SWR is flat. Also, the forward meter goes to zero. I don’t know if this is caused by a marginally defective op-amp, or just a cold solder joint. I didn’t want to send it back to MFJ and wait months for warranty service, so I fixed it, finally, by simply shielding the meter board. But first I tried re-routing the cable going from the directional coupler to the meter board.

This is a difficult problem to troubleshoot because it only happens when the tuner is assembled, and it must be disassembled to get to the meter board.

The First Attempt


At about 250 watts,


WHACK! the reflected meter pegs out. I switched to another band where the capacitor tunes at a lower number, and indeed, the problem did not occur, proving that it’s the proximity of the capacitor’s plates to the meter board.


Removing the giant capacitor, I see the cable runs right under it, so I thought maybe re-routing it might cure the problem. Also, I noticed that the leads (white and black wires) run unshielded all the way across the board. UGH!


Remove the antenna switch shaft, then the meter board.


I decided to move the cable to the other edge of the board and shorten its unshielded leads.


There’s a handy ground point there for the shield.


I cut several inches of leads off and re-soldered them to the board.


Then I mounted the board and routed the cable along the side of the box, and taped it down with foil tape, so it wouldn’t be under the capacitor.


put the shaft and bread-slicer back in, solder the input wire and capacitors to it, aaaaaannnnnnd…


WHACK! No change. Wasted time and effort. RF must be getting directly into a component, not by the wires.

The Second Attempt

Discouraged, I put it back together and left it for awhile. Then a couple weeks later I decided to try again. It occurred to me that perhaps shielding the whole board might be effective.


Yes, it would be prettier with folded sheet aluminum, but I didn’t have any (much less a sheet metal brake), so I just used foil tape. I left the paper on the back to insulate it from the components.



It was quite difficult to re-mount the board with the cover in the way. I managed to get one corner under a stand-off, and the opposite corner under the leg of the capacitor. Hopefully, two grounds will be enough.


TA-DAAAAA! It works! There’s 800 with no whack.

MFJ really should shield that meter board… And fix their counter bracket bending machine!

Replace the LM324N Quad Op-Amp Chip

In June of 2020 I had a very close lightning strike, and even though the antennas were disconnected, and the 989's meter power was off, the impulse blew the meter's op-amp chip. Symptom: Turn on the meter, it lights up, but does not respond to RF. Looking at the circuit in the manual, it had to be either the diodes on the directional coupler board, or the LM324 quad op-amp IC, or a couple of transitors, on the meter board.

Using a DMM I first determined that the directional coupler was indeed putting out a little DC with RF power applied through the 989D, so the diodes are okay, and it had to be the meter board. It could be either the chip or the transistors.


You can either get a couple copies of the exact parts you might need (and pay too much shipping) at an electronics supplier, or just get assortments on Amazon, for about the same price.


For the IC assortment, make sure it has the actual part number you need (in this case, the LM324), and sockets. If you ever go to all the trouble of removing a soldered IC, you'll definitely want to replace it with a socket!


For some reason, I thought I'd have to unsolder the chip, so I removed the big capacitor, shaft, meter, and board.


D'OH! It's already socketed! I didn't need to remove the board after all. Popping in a new LM324 fixed it.

So, if your meters simply die, start by testing for DC from the directional coupler board (with a little RF applied) to verify the diodes. Then replace the chip, then test. If that fixes it, good. If not, remove the meter and board and replace the transistors, which are the very common 2N3904.

What’s Next?

In the next article, I will attempt to document the building of my latest antenna, the 80 Meter Doublet! I built it super-heavy-duty and put it up to stay. Two continuous runs of #10 wire form the antenna and ~150 feet of home-brew 4-inch ladder line.

73, — KV5R

16 thoughts on “MFJ-989D Tuner
  1. Just bought a new (11/24/2023) MFJ-989D from HRO but did open it up before putting it to use just to check everything out. I was very fortunate to get one that appears to be perfect! Gear alignment for the turns counter is perfect, and they have also put a heavy aluminum shield over the meter board that is bolted to the chassis and custom bent to fit the board with holes in place to access the calibration adjustments. Only thing I noticed which is no big deal for me is the holding bracket for the 9 volt battery which I would never use anyway. One other thing was it had a small, round knob for the antenna selector that was hard for my arthritic fingers to get a good grip on, so I found a winged, chicken head knob in my junkbox and replaced it and not it is easy to turn.

  2. Thank You Harold,

    I’m considering buying a 989D tuner and found this. Very valuable and well written article. Now I know what I’m up to if I end up having one.
    73, de Jay OH1XYZ

  3. I had trouble on 10m as you apparently did. Sweeping the balun with a NanoVNA showed that it was essentially a low-pass filter that starts becoming a near-dead-short just above 25MHz. I suspect the balun has the wrong core material or too many turns for 10m. With an external “ugly” balun, it works just fine. I’m going to replace/rebuild the internal balun to correct this problem.

  4. SALUT JAI MODIFIE MON MFJ QUI ME DONNAIT DE MAUVAIS RESULTAT EN UN CONDENSATEUR DIRECT ET UN CONSENDATEUR ALA MASSE ET CELA FONCTIONNE TRES BIEN MAIS J AI PAS PU ME METTRE A MOFIFIER MON CIRCUIT DE SWR QUE JE VEUX PAS D ALIMENTATION 12V OU PILE JE VEUX DIRECT

  5. To day 02 / 02 / 2022 i make solution for my 989D versa Tuner, I install a aluminum paper over meter Board is all OK, no RF in the meters. Thanks Steve, K5DJH my name Santos, Florida, kD2SM 73s.

  6. Hello! Have your ever looked at an article by PA0FRI named “MFJ-989D VERSA TUNER V conversion to S-Match”? I use s-match technice quite often and it works Wall. Regards sm7cbs Tore Sandstrom

    • Yes, I have seen the s-match mod. I just don’t need it here, using the coax 1 output to external choke/balun (Balun Designs 1171) it works well.
      73, –kv5r

    • FB! Yeah, it’s a shame mfj can’t seem to go one extra step and put a 4″ square aluminum sheet with 4 bends in it and shield that board…

  7. Bought my MFJ-989D from a SK’s estate sale. And, pretty much like everything else MFJ, everything in them are fixable and modifiable. That roller inductor counter? Standard counter used in cassette decks for decades, and they were never dead-nuts perfect, but always at least ball park. So, can work with it. The roller inductor handle bound up on me 1st try. Taking the tuner apart revealed that the lube at the end brass bushing dried out. So, I did all friction axial points with Dow Corning Silicon Vacuum Grease. Rolls nice & smooth now. (Never put that grease on the inductor itself, or the contact bearing.) That 12VDC switching wallwart for meter powering? Ditch it! Grab a 12VDC transformered wallwart from a junk store and use an LM7812 +12VDC regulator after that. Now, you have a meter DC power supply that won’t add RF spurs to your RX signals. Neat site!!!! 73!!!!!

    • Good points on the grease, and the power supply. I just connected mine to the main 12V system, which is a Samlex 1223 (with added type-77 toroids) followed by a lawnmower battery. I came off the main supply with a 5A fused line to a 12-point barrier strip for all the little giblets.
      73, –KV5R

  8. I’ve had many MFJ products with absolutely no problems at all, but I am aware that some get through the production line with problems that should never have gotten to the customer. I think, for the most part, MFJ products are great value for the money, but there are better made products (like Palstar tuners, for instance) that cost a lot more. LDG seems to be about the same. Most are just fine, but some get through with problems and end up in an unhappy customer’s hands, like the one LDG auto tuner I read about that caught on fire on a field day operation. (I’m sure some unpleasant experiences have happened with MFJ products, too.) I’ll stand by MFJ for now, at least, but I won’t be surprised if I end up with a product that has a bad solder joint, sheet metal holes that don’t line up, or some other defect. If I had the money, I would buy better quality products, but I am nearing retirement and have to get used to spending less now. 🙂

    It seems that some of their products have more consistent quality than others. If production is considered when designing a product. I think the 993B tuner is one of those and it gets decent ratings (but not perfect).

    Donald

    • Yes, I agree with you completely. I too have several MFJ products, including a 993, which is still working fine after about 16 years.
      This article was written to help others that might have similar problems with the 989D, not intended to bash MFJ. And like you, I have to consider the cost; radio accessories have gotten way too expensive in recent years (along with everything else), and I’m glad MFJ tuners are still pretty reasonable.
      73, –KV5R

  9. I had the same problem with the reflected wattmeter you did. I put in a ticket with MFJ and never heard back from them. It only happened with high power using balanced line. I thought the RF was getting in to the electronic wattmeter. I put caps across the power leads on the board but that didn’t work. I was glad to see your article. I fix the problem in the same way.

    Thank you very much for putting the your article on the internet.

    73,
    Steve
    K5DJH

Leave a Reply

Your email address will not be published. Required fields are marked *