DTV LNB Mod; Ku Conversion, Part One

[CharredPC]

Fist off the standard disclaimer; this is a little tutorial on how to possibly destroy your perfectly good (if all too common) DirecTV LNB! Do not attempt this unless... well, you know- you're over 11, have your parents and/or kids' permission, are Vulcan and have two hearts... Whatever.

Big thanks to sidha for giving the original method. Following his suggestions will undoubtably work better. Consider this the way of going about things if you have none of the proper tools or parts

STEP ONE:
Find an old DirecTV LNB.
http://www.dwarchive.net/mod/mod1.jpg
Here's my unsuspecting old dish, hanging out behind the house. You're goin' DOWN, LNB! Well, down in frequency anyway..

The LNB off mine happens to be the "easy-to-mod" kind:
http://www.dwarchive.net/mod/mod2.jpg
Namely: California Amplifier P/N: 15057 LOT NO: 02385
I highly recommend you seek out a similar model, as most others are purposely tamper-resistant, and a REAL pain to mod, or even take apart.

STEP TWO:
Crack it open. Just prying with a screwdriver should do it.
http://www.dwarchive.net/mod/mod3.jpg
Don't worry about being a bit rough with the screwdriver; no vital parts exposed yet.

STEP THREE:
Remove the feedhorn cap.
http://www.dwarchive.net/mod/mod4.jpg
As you can see from the picture, now you can see the teflon (looks like plastic to me) "wedge" glued in the throat of the feedhorn. We want this out to convert from circular to linear polarization.
WARNING! Attempting to pull it out with pliers results in this:
http://www.dwarchive.net/mod/mod9.jpg
The wedge is fairly brittle and easily pulls apart that way. Keep reading for a better method of extraction...

STEP FOUR:
Pry the back cover off the LNB.
http://www.dwarchive.net/mod/mod5.jpg
The cover is held in place with some rubbery sealant that is easily scraped away on my model. It's there to keep the weather out, not you

Now we can see the innards of the beast:
http://www.dwarchive.net/mod/mod6.jpg
Note the tuning screw. A vey simple mod is to adjust this screw to skew your frequency coverage by up to 23Mhz, says sidha; then the very uppermost Ku transponders might be findable with a blind search.

STEP FIVE:
If you're like me and don't have the 'right' tool to loosen the backplate screws, I suggest using some pliers and carefully giving each a half-twist or two. Once loosened that much, they turn smoothly and can easily be unscrewed/removed with your fingers. Once done, we now have this:
http://www.dwarchive.net/mod/mod7.jpg
Note the ceramic "puck" under where the tuning screw was. To lower our frequency, we need to add to this puck. A reverse mod some may be familar with is sanding this puck down, raising the freqency, and converting this for other uses (as discussed here ).

STEP SIX:
Remove the teflon wedge. The best way seems so be carefully using a flathead screwdriver to break the wedge's glue seal, then (again, carefully!) from the reverse side, using a flathead screwdriver and a hammer, driving the wedge out, like so:
http://www.dwarchive.net/mod/mod8.jpg
Yes, I said be very careful, but if you do crack it up a bit rest easy knowing mine looks like a dog chewed on it, and there is about 5-10% physically missing, but it still works at least as well as it did previously.

STEP SEVEN:
Add ceramic to the 'puck' to lower the frequency.
Sidha tells us to use a 'donor' LNB and cut/shave bits off to add to this modded one. Having no donor LNB, I looked around my junk pile for some suitable ceramic. Hrmmm...
http://www.dwarchive.net/mod/mod10.jpg
Yes, here's what I was looking for:
http://www.dwarchive.net/mod/mod11.jpg
I'm sure using another LNB for donor parts is much more suitable, but this seemed to do the job Now, break it!
http://www.dwarchive.net/mod/mod12.jpg
Try and get some decent-sized chunks. I also made sure the chunks had no metal in them, just ceramic; it might not matter in small quantity, but I wasn't taking chances.

STEP EIGHT:
Add the ceramic bits to your 'puck.'
Sidha recommends superglue or the like. Not being able to find any in the house at the time, I tried to think of what other substance I had that could 1) go on in a very thin layer and 2) be sticky enough to keep the bits in place. I finally found some thermal paste that seemed to suit nicely. Again, I trust in sidha that superlue or epoxy are better choices. With enough of a beating, I imagine the paste could loose its' grip on the ceramic.
Okay, so, I coated the puck a bit:
http://www.dwarchive.net/mod/mod13.jpg
... and then added some ceramic chips:
http://www.dwarchive.net/mod/mod14.jpg
Your goal should be about 30% more cermaic added on to the 'puck.' I did a lot less this time, becuase I'm lazy. Actually, I wanted to finish writing this up and get things working before 4am Try and add as much as you can around the edges, and only thin peices on top... there's lots of space around it, but not much height-room to add.

STEP NINE:
Reassemble the LNB. This is (sort of) where Part One of the tutorial ends. I can offer some helpful hints, but those with recievers capable of blind scanning should be pretty well off now; SOME part of the Ku band should be within your reach if you added enough ceramic. Fine tuning with the set screw can be done to adjust your frequency range- also, when putting everything back together, watch that the set screw doesn't crush your now-higher 'puck.' Here is a little proof-of-concept I've developed from my non-blind-scanning Twinhan card:

I've had my dish locked on Echo7 the entire time I did this mod. Not as easy as it sounds, I practically have to hang off the dish to remove/reinstall the LNB First, I put the wedge back in to restore circular polarization. I fired up ProgDVB, and tried pulling up Echo7:
http://www.dwarchive.net/mod/mod15.jpg

As you can see, I got nothin'. Why would this be? We've changed the LO on the LNB! So I'm way off frequency trying to pull in the transponders. I didn't add much ceramic to the puck, so I shouldn't be that far off. Adjusting bit by bit, and checking the same transponder each time, I finally arrived at:
http://www.dwarchive.net/mod/mod16.jpg
Yes, the LO is now set 180Mhz below normal. And we get:
http://www.dwarchive.net/mod/mod17.jpg
Incedentally, this is the best quality I've ever gotton off this dish. Actually, I've never had more quality than signal... weird, huh?

The bottom line is, I now have a functioning new LO! With more ceramic, it would drop lower and lower; with time, patience, and some tuning screw adjustments I can achieve 10750. With the wedge removed, we have a Ku-band capable DirecTV dish! Yay!

I'll continue to work on this. Hopefully sidha will offer more advice, and I'll write up a Part Two when I successfully lock into a linear polarized Ku bird. In the meantime, I hope his little guide helps everyone! It's a fun project anyway, and the parts are sure easy to come by

CharredPC

 

[timmy1376]

So basically, it's trial and error to get it to the frequency of Ku band stuff? Would this work with a dual lnb? how about an OLD dish network lnb?

 

[CharredPC]

Part 1A; Alignment of the LNB

So basically, it's trial and error to get it to the frequency of Ku band stuff?

It's trial and error, yes.

Sidha may correct me on this (please do if I'm way off base!) but here's my basic formula on re-aligning your LNB after adding ceramic:

-Point your dish (if it isn't already) at a known bird. I used Echo7 (leaving the teflon wedge in place to keep cirular polarization so I could actually recieve E7).

-Check one of the lower transponders for a quality reading.

-Slowly change the LO frequency- about 2-5Mhz at a time. This part is trial and error. You have to actually re-pull-up the transponder each time on my Twinhan; it's doesn't auto-change after a LO adjustment.

-Once you find your LO, quality will be restored. For the above experiment, I had some quality return at 11072, but 11070 let me recieve all E7 transponders again.

If you don't mind running out to the dish a few times, a slightly easier way to do this (which I've successfully done) is performing the above LO-adjustment method after adding only a little ceramic. Once I found the new LO, I add a little more ceramic, and locate the LO again, a bit further down. Keep going adding bit by bit, until you're near the magic number of 10750. Once you're within about 10-15Mhz, you may be able to use the tuner set screw to fine-tune to "perfectly" lock in a 10750 LO.

Now, at this point I'm slightly unclear (and hope sidha will help out) about the settings. I believe we now have a standard Ku LNB, so in ProgDVB I should alter the "circular" setting to "Ku-Band." That, and removing the teflon wedge completes the modification. We should now have a linear Ku dish. If you know where a good Ku bird is, aim at that and check your transponders. Again, if you get only half the transponders (or only a few) but know you are pointed correctly, try changing the LO by a Mhz or two (are carefully adjusting the tuner set screw slightly). Careful not to crush your larger 'puck' with the tuning screw! Using the above method, I believe you will remain on-frequency and not need a blind scanning reciever to verify the results. If I could map out where a linear Ku bird is located, I could complete testing myself; I plan on doing so soon, maybe after I have my Winegard dish next week and can pinpoint the proper angle/elevation for certain birds.

Would this work with a dual lnb? how about an OLD dish network lnb?

According to sidhad, the mod is basically the same on other LNB versions, the main differences being in the difficulty. Some LNB's have heavy-duty sealant to break through. Some have the wedge physically molded in the throat, and require a file to cut down, instead of having an easy seperate wedge to remove. My version had screws holding the backplate on; some are held in place with epoxy, and must be carefully pried off. I don't have several versions to compare at the moment; I welcome others to post pictures of their LNB's and model numbers.

As far as a dual LNB, I have no idea... yet. If you mean the versions with twin LNB's on one dish, I see no reason for this mod not to work fine. It may even be possible to mod one LNB and not the other, creating a Ku/DSS dish like some others have done by swapping out a LNB. I'm hoping when a few others try this mod, we'll get some more pictures and model-specific tips for those who don't have this particular version.

CharredPC

 

[sidha]

snip// A very simple mod is to adjust this screw to skew your frequency coverage by up to 23Mhz, says sidha; then the very uppermost Ku transponders might be findable with a blind search.


I lowered the LO frequency, but I could have raised it by about the same amount. +/- 23 MHz was about the limit on one particular DTV lnbf model; on a random DN model I lowered the LO 200 MHz by adding additional ceramic to the puck. Modifying the puck will avoid the loading down effect on the oscillator caused by attempting to lower the frequency by adjusting the screw too close to the puck. My DTV lnbf has an easily removable back plate with O ring; all screws used for external and internal covers are standard Phillips. The DRO adjustment screw slug is almost the diameter of the puck (thus a wider adjustment range), but not externally accessible. The (DN) lnbf adjustment screw is accessible from the outside, but provides a rather limited range of adjustment. Also, the DN unit required the removal of a metallic fin inside the throat of the feedhorn to allow linear reception. The inner wall of the feedhorn should be a continuous smooth cylinder when the job is done. Do not remove any small metal fins that may be inside the inner shielding cover, only file or ream the throat area of the feedhorn. Begin this procedure only after the vulnerable circuit board is removed and safe from metal debris. (Need I mention static protection?) I epoxied a thin puck fragment to the top of the puck, increasing it's height by about 15%; the LO dropped about 200 MHz. Your mileage may vary.

If possible try to obtain ceramic material from another dielectric puck used in a similar microwave application, like another lnbf, to avoid a thermal drifting LO. I ran my modified lnbf's all winter and rescanned periodically for LO drift that would be reflected in the IF transponder frequency. Both DRO modified lnbfs performed very well in all respects. I modded another DN lnbf that also required filing the circular polarizing fin from the feedhorn throat. That unit also has a convenient externally accessible adjustment screw, but the screw-less thin metal back cover is semi-permanently silicone sealed, like the sardine can you opened. I added 30% of additional ceramic to the puck in about three attempts, until the LO reached nearly 10750 MHz. The fine tune adjustment screw did the rest. Keep the screw at mid position while adding ceramic, and try to get within 10 MHz before using it. This lnbf has been in use for over six months; it's stable with no problems or lack of performance. In fact, I'm impressed.

The modded lnbf is now a Ku linear lnbf, so program the parameters to recognize it as such. There is one little unknown concerning the voltage switched polarity giving opposite results due to the possibility of reversed antenna positions in the waveguide, and the V skewing effects mentioned below. You picked a tough lnbf for the project. At least on Echo 7 you can prove exactly where you are by accessing a music or free video channel to compute frequency deviation and extrapolate results.

I'm without a way to link pictures to this forum, but I have some internal shots if you're interested. Also, your photo of the circuit board reveals etched antennas skewed in a V shape (to horizontal) instead of L shape. Either one will need to be horizontal and matched to the skew for the satellite. Can you guess which one will be right or left? The dual antennas inside my lnbf's are positioned standard Ku linear, and so the lnbf's mount normally on my dishes. I bet you will need to rotate the dish to correct the skewing for that lnbf. There are so many variations to be found in different brands and models. BTW, I've begun to modify a Dish Pro head with one shared DRO; two birds with one puck, but, alas, no Ku & DSS.


more, later...

-sidha

 

[CharredPC]

My DTV lnbf has an easily removable back plate with O ring; all screws used for external and internal covers are standard Phillips.

That must be nice! Mine has little hex-nuts; one would need a very very small socket tool to removed them 'correctly.' If they were just a bit larger I'd use a dremel tool to cut flathead-style grooves into each head, as I have on other projects in the past.

If possible try to obtain ceramic material from another dielectric puck used in a similar microwave application, like another lnbf, to avoid a thermal drifting LO.

I agree.
That being said, I've monitored my home-rigged cermaic mod during at least a 40 degree temperature change, with no visibile variance. Of course, as sidha said, your mileage may vary.

The modded lnbf is now a Ku linear lnbf, so program the parameters to recognize it as such. There is one little unknown concerning the voltage switched polarity giving opposite results due to the possibility of reversed antenna positions in the waveguide, and the V skewing effects mentioned below. You picked a tough lnbf for the project. At least on Echo 7 you can prove exactly where you are by accessing a music or free video channel to compute frequency deviation and extrapolate results.

Tough from it having a "V" element design, as opposed to an "L," you mean? Yes, I thought of that. I'm trying to come up with a good way of skewing the lnb while keeping it proper length and orentation from the reflector dish, or else skewing the whole dish. Needless to say, the mount was never meant for either possibility.

I'm without a way to link pictures to this forum, but I have some internal shots if you're interested.

charredpc@msn.com ; I'll gladly post/host any pictures you have.

BTW, I've begun to modify a Dish Pro head with one shared DRO; two birds with one puck, but, alas, no Ku & DSS.

Keep us posted! If I can gather enough LNB data, I'll toss a reference guide type page up that should be helpful to all.... what LNB's work best, the P/N's, and specific differences in the design.

So, overall, do I have your theory down? My plan is to keep the dish using circular polarization, and locked onto Echo7 while doing adjustments. Once my LO is the proper 10750, I should have a properly calibrated LNB. I can then remove the wedge, skew the dish, set the LNB to Ku linear polarization, and start scanning for Ku transponders.

CharredPC

 

[ken2400]

I hope to work on this mod soon BUT will be tied up this week. I hope to bring some LNBF to work and take them apart. I did one once this year BUT did not photo it. I hope to Tue or Wed just to show what I have to work with.

I am REAL that all my work as inspired others to add on to it.

That must be nice! Mine has little hex-nuts; one would need a very very small socket tool to removed them 'correctly.' If they were just a bit larger I'd use a dremel tool to cut flathead-style grooves into each head, as I have on other projects in the past.



I agree.
That being said, I've monitored my home-rigged cermaic mod during at least a 40 degree temperature change, with no visibile variance. Of course, as sidha said, your mileage may vary.



Tough from it having a "V" element design, as opposed to an "L," you mean? Yes, I thought of that. I'm trying to come up with a good way of skewing the lnb while keeping it proper length and orentation from the reflector dish, or else skewing the whole dish. Needless to say, the mount was never meant for either possibility.



charredpc@msn.com ; I'll gladly post/host any pictures you have.



Keep us posted! If I can gather enough LNB data, I'll toss a reference guide type page up that should be helpful to all.... what LNB's work best, the P/N's, and specific differences in the design.

So, overall, do I have your theory down? My plan is to keep the dish using circular polarization, and locked onto Echo7 while doing adjustments. Once my LO is the proper 10750, I should have a properly calibrated LNB. I can then remove the wedge, skew the dish, set the LNB to Ku linear polarization, and start scanning for Ku transponders.

CharredPC

 

[CharredPC]

From sidha

DBS lnbfs that I modified for Ku band.
(outer shells removed)
Notice external set-screw adjust on DN model.

http://www.dwarchive.net/sidha1.jpg
http://www.dwarchive.net/sidha2.jpg

DN lnbf
Notice the O ring seal and antenna orientation.
Great design.

http://www.dwarchive.net/sidha3.jpg

DTV lnbf
Lowered 23 MHz by set-screw alone.

http://www.dwarchive.net/sidha4.jpg
http://www.dwarchive.net/sidha5.jpg

DN model
Thin slice of ceramic transplanted from donor lnbf.
A very thin layer of epoxy was used.
LO frequency change: -200 MHz.

http://www.dwarchive.net/sidha6.jpg

Done

Also, an update on my own project:
After snapping off one of the antenna inside my LNB, I wanted to get some parts. I picked up another LNB at the thrift shop for $3 (includes the useless DTV dish) and here's what I got:

http://www.dwarchive.net/lnbtwo1.jpg

... and inside:

http://www.dwarchive.net/lnbtwo2.jpg

Well, at least the screws were phillips, but the puck is protected by a soldered-on housing! How annoying! I removed the element/antenna cover;

http://www.dwarchive.net/lnbtwo3.jpg

Uhg! A variation on the "V" design again! ... and I was hoping for an 'L.'
I decided this could be my donor LNB, and I'd rebuild the original. Pay attention to sidha when he warns you of the delicate circuitboard!
I then opened the puck 'tin,' using my handy dremmel tool, and removed all the proper ceramic I'd need to mod my initial LNB.

http://www.dwarchive.net/lnbtwo4.jpg

Now I just had to figure out how to fix my broken-off antenna element. Hrm. Actually, the other one looks kind of cracked too... well, ok. Snap! I'll start fresh. Back to the parts box I go...

http://www.dwarchive.net/lnbtwo5.jpg

That's what I need! A little unsoldering, a bit of bending, trimming, and voila!

http://www.dwarchive.net/lnbtwo6.jpg

My poor abused LNB springs back to life! The replacement antenna work great! Too bad I can't reposition them into the proper 'L' shape (yes, I tried, but the elements are very length-dependant and stop working if made longer). I added chips from the donor's puck until my LO dropped down to 10910. I found that using just the one Echo transponder that contains Angel, and verifying that the proper channel was recieved (instead of checking multiple transponders, and finding "overlapped" signals) helped a great deal. However, dropping below 10910 pulls Echo out of range, so I can no longer test/find the new LO. Hrm..

About this time, I was told that my Winegard 31", SG2100, Universal LNB are showing up tommorrow, and this project didn't seem quite as important as running the cable to the roof, planning the perfect mounting spot, and reading the SG2100 manual and setup guide

 

[ken2400]

I found a LNBF to mod. I started pulling it apart BUT need to find my security bits.

My question is what have people picked up with there Modified LNBF?

Thanks

 

[sidha]

I found a LNBF to mod. I started pulling it apart BUT need to find my security bits.

My question is what have people picked up with there Modified LNBF?

Thanks

I have four DBS LNBF's with re-tuned LO's now oscillating at 10,750 MHz. I'm using one on a Primestar dish receiving all of A5's Ku transponders. As far as performance on the original matching 18" dish, I can solidly watch People's Network on AMC4 with a quality level of 12 and 3ABN booms in at +60. Using a 20" (Dish 500), People's Network is well into the yellow and 3ABN is in the 80's. Also, some stronger transponders on A5 are watchable. With these results and the growing surplus of free DN & DTV LNBF's, I doubt if I'll ever buy an Ku LNBF, new or used. Anyone with some skill and ambition should give this project a try because the reward far exceeds the cost. Having a blind search receiver helps to guide the DRO adjustment process, but it's not essential if you're very patient and committed to reaching the goal. (No pun intended, Hockey fans.)

-sidha

 

[ken2400]

A5? is that IA5? 97W? Have you tried G10R 123W?

Thanks

I have four DBS LNBF's with re-tuned LO's now oscillating at 10,750 MHz. I'm using one on a Primestar dish receiving all of A5's Ku transponders. As far as performance on the original matching 18" dish, I can solidly watch People's Network on AMC4 with a quality level of 12 and 3ABN booms in at +60. Using a 20" (Dish 500), People's Network is well into the yellow and 3ABN is in the 80's. Also, some stronger transponders on A5 are watchable. With these results and the growing surplus of free DN & DTV LNBF's, I doubt if I'll ever buy an Ku LNBF, new or used. Anyone with some skill and ambition should give this project a try because the reward far exceeds the cost. Having a blind search receiver helps to guide the DRO adjustment process, but it's not essential if you're very patient and committed to reaching the goal. (No pun intended, Hockey fans.)

-sidha

 

[sidha]

A5? is that IA5? 97W? Have you tried G10R 123W?

Thanks

A5 = Intelsat Americas 5 = IA5. I haven't tried G10R because it's not a visible target from the dish farm on my rear deck -- due to leaves on trees. However, I have a Primestar dish installed at the west side of the house to get G10R and a couple other birds. I could set-up a small 18 or 20" dish with modded LNBF at that location and see what happens, when the rain stops, eventually.

sidha

 

[CharredPC]

I'm actually using FrankenLNB (Cascade's term for it) at this moment. The twisted, totally wrong fact is it's the best working LNB in the house, and I've got a DirecPC and new Universal LNB sitting here on my desk Is that messed up or what?

Oh, I solved the the "V" design problem in 2 seconds ith this new Winegard dish; just mounted the lnb 90 degrees to the side. Works great!

CharredPC

 

[sidha]

Congratulations!

As far as I know, you are the only person besides me to prove that it can be done, and without the help of blind scanning. Now you know how well these LNBF's perform, and you'll want more. One important point: The tuning screw must be at a safe distance from the dielectric puck, or you will over dampen the oscillator and severly reduce the IF signal output. Instead use sufficient ceramic when lowering the LO frequency so that the adjustment screw will not cause negative effects. Monitoring signal level while turning the screw clockwise towards the puck will quickly reveal the optimum range for peak output. This is where a signal level meter is the tool to have outside at the dish.

sidha

 

[techno935]

RE-activating the thread since this was just recently brought up. CharredPC, do you have those pictures available?

 

[Tron]

WOW... how were you able to respond to this thread? Usually, the forum software doesn't let anyone respond to a thread this old...

 

[gabshere]

do you have those pictures available?

I would also like to see them

 

[techno935]

I re-activated the thread to allow a previous poster who made a similar thread recently to hopefully get some answers. Maybe Ice can merge the threads. I did try to modify a DTV32 Eagle Aspen LNB for linear. No such luck and no tuning screw. I did however explore the innards!

If the OP can come around and show us the pictures or find em again, it would be great, then the thread can become a sticky in a future "mods" (modifications) section.

We should have that shouldn't we huh? That would be a nice section. Safe hardware mods that don't involve theft of any kind. Even mods on dishes. (Like mine!)

P.S. To re-activate an old thread, you have to click some check box or something. I hate to do it but I don't like making new threads on something that was previously posted ya know.

 

[pedrogarcia]

You know when I started reading this thread I thought why would anyone want to bother to do this when modern linear LNBs have better noise specs and are realtively cheap.
Then I read on and started looking at my scrap box and there were some old Direct TV lnbs, why the heck did I even keep them. Of course, it was just in case I might use them for some hairbrain project. Well 2 are now apart and
about to be modded. I'll let you know how I get on.
Thanks Guys - I think

 

[sidha]

You know when I started reading this thread I thought why would anyone want to bother to do this when modern linear LNBs have better noise specs and are realtively cheap.
Then I read on and started looking at my scrap box and there were some old Direct TV lnbs, why the heck did I even keep them. Of course, it was just in case I might use them for some hairbrain project. Well 2 are now apart and
about to be modded. I'll let you know how I get on.
Thanks Guys - I think

A revival. Great to see renewed interest in this project. A chance to see what other experimenters might use for technique, and their results. I wanted to keep my project down to using tools that everyone might already have -- basically, an adventurous brain. CharredPC hosted my series of conversion pictures at the time. I [should] still have all of them somewhere on at least two HD's. I hope to repost them, soon. And if CharredPC notices the activity or PM, he might be able to resend his photos too.


-sidha

 

[pedrogarcia]

Well I have managed to convert 1 LNB without resorting specialized equipment although I was really tempted with the LO setting. As I am in Europe at the moment I set my dish on Arabsat 26 east which has a massive signal and was pleasantly surprised with the quality, on a weaker sat 12w I still got good quality. It suffered some frquency drift at the top of the scale and could not reach all the european Ku higher frequencies but it was a really worthwhile project.
My other LNB was a complete failure due to my complacency at winning first time and I mucked up the IF stage.
Back in the States I will have another go.

In case you were wondering why I took circular LNBs to europe in the first place it was to get the English football and hunting and fishing FTA channels on 36 east Russian Satellite