Canadian TV, Computing and Home Theatre Forums banner

CM4228HD Hardware Hacks

236K views 457 replies 89 participants last post by  wkrwu 
#1 ·
I modded my CM4228HD as per Ken Nist's excellent article. I removed the hex screws holding the rods with the plastic balun box. I added a separate balun to each of the two antenna sections and a combiner. Measured with a laptop at the antenna using WinTV's Digital Signal Quality Meter I saw an increase in SNR from 27 to 30. I receive 4 digital stations, all broadcast from Camp Fortune at about 30Km north. The analog VHF Hi stations are also very clear; the VHF lo are a bit fuzzy. I'm about 5Km from Herbert's Corners, the other local broadcast tower; it's to the south. The UHF analogue stations and Ch 11 are watchable, but a little fuzzy. They are being received from the back of the antenna! The whole rig is in the garage rafters on the north side of the house.



I then took a CM4221HD and removed the bowtie assembly. I mounted that to the back of the CM4228HD; I'll call it a CM4229HD. I added another combiner that merges this new antenna section with the output of the first combiner of the modded CM4228HD. I saw no loss in the SNR meter and the picture quality of the VHF-Hi & UHF analog stations in the north direction were all still excellent. But now the UHF and Ch11 from the southerly Herbert's Corners are also very good. There is a slight multi-path ghosting on some of them, but still watchable.



Next, I disconnected the laptop from the antenna and replaced it with the 20 metre cable that leads to the VHF/UHF input of an Eagle Aspen S-4180-GX+ multiswitch. It mixes the VHF/UHF signal with the input from the satellite LNAs and distributes it to TVs and radios around the house. The loss for VHF/UHF through the multi-switch is 5dB. The signal at the TV outlets was pretty degraded. I couldn't get 2 of the digital stations from the north; most of the analog stations were very fuzzy.

Next, I put a CM7777 preamp at the antenna. This seemed to overwhelm the input of the tuner. I added a cheapo splitter between the preamp power supply and the multiswitch as a 4dB attenuator. That seemed to clear things up pretty well.

So, I'll probably leave everything this way until Canada's digital switchover and decide if it's worth the effort of putting this rig outside.
 
See less See more
2
#6 ·
Thanks, I will definitively try this. Lately, I just tight wrap the balun inch back and lost uhf 43. It was no fun to examine the balun closer in the attic.

Techluvr,

I guess you have remove the 2 original horizontal rods where the square balun is connected to? I read somewhere that it is impossible to pop open the balun.
 
#7 ·
Techluvr,

I guess you have remove the 2 original horizontal rods where the square balun is connected to? I read somewhere that it is impossible to pop open the balun.
Yes, it's very easy to do. It can be removed with just a hex-nut driver. I was careful not to bend the rods in case I had to put it back together. As it turns out I won't need to do that because I think the modified antenna works better. Of course that's just my subjective observation. I'm going to try to back that up with some real numbers this week-end. I'll post my results here.
 
#11 · (Edited by Moderator)
Modified 4228HD... Some real world numbers

My 'real world' is a CM4228HD mounted in the rafters of the garage. The mast is a 1" copper pipe that I can rotate 360 degrees. I am located between two broadcast towers almost 180 degrees apart. One is north ~30KM; one is south ~ 5KM.

I took two sets of measurements from every UHF channel from both towers; one set with the original CM4228HD and one set with the centre balun replaced by 2 individual baluns (-2.5dB ea) and a combiner (-.5dB). The <10M cable losses and any other incidental connector losses are the same for both setups, so I ignore them. I took an additional set of measurements with a CM4221HD for only the south facing stations.

All numbers are dBm rounded to the nearest whole number.
Code:
    Ch        Orig    Mod   4221HD
    [B]14[/B]        -33     -34     -29
    [B]20[/B]        -58     -58
    [B]22[/B]        -52     -56
    [B]24[/B]        -44     -42
    [B]25[/B]        -50     -48
    [B]30[/B]        -47     -45
    [B]34[/B]        -51     -53
    [B]40[/B]        -57     -58
    [B]43[/B]        -50     -48     -44
    [B]60[/B]        -60     -54     -53
    [B]65 [/B]       -57     -48     -48
Not fantastic; but for the most part, the modification is an improvement over the original. The 4221HD did just as well if not slightly better. Maybe if I have time, I'll swing the 4221HD north and see how it does on those stations.
 
#15 ·
It's an Agilent E4402B. It's an 'economy' model; that's why they let me take it. :rolleyes:

The procedure ( I'm sure that's the next question )...

Set the frequency to capture a contiguous group of stations from the target broadcast tower. Set scan rate of 500mS with trace mode set to 'erase/write'. Place marker on a center channel. Go up ladder and slowly rotate antenna while reluctant helper (wife) reads out dBm @ marker. Stop when highest figure obtained. Dismiss helper. Set freq span to entire UHF band; trace mode to 'max hold'. Wait for trace to stabilize at max values. Move marker to each peak and check if the frequency is a channel from the target broadcast tower. Record ( paper & pencil ) the dBm reading. Repeat from start with other broadcast tower.

Note: For each direction there was significant signal detected from the stations in the opposite direction. I'm not sure if this is off the back of the antenna or from multipath reflections. I do see ghost images with a large enough displacement that makes me believe it's reflections from distant objects.
 
#16 · (Edited by Moderator)
Code:
    Ch        Orig    Mod   4221HD
    [B]14[/B]        -33     -34     -29
    [B]20[/B]        -58     -58
    [B]22[/B]        -52     -56
    [B]24[/B]        -44     -42
    [B]25[/B]        -50     -48
    [B]30[/B]        -47     -45
    [B]34[/B]        -51     -53
    [B]40[/B]        -57     -58
    [B]43[/B]        -50     -48     -44
    [B]60[/B]        -60     -54     -53
    [B]65 [/B]       -57     -48     -48
techluvr, you don't get CITS channel 32 from Herberts Corners? I know it's the weakest station, so maybe it's buried in the noise on your analyzer.

How do your measured results compare with TVFool Longley-Rice predictions (after accounting for the antenna gain and cable loss?
 
#18 ·
techluvr, you don't get CITS channel 32 from Herberts Corners? I know it's the weakest station, so maybe it's buried in the noise on your analyzer.
It was more of a procedural error. Ch 32 is the only HC channel between ch 20 & 40. I didn't realize I skipped over it until I had already modded the antenna. I didn't want to go back and un-mod just for one more reading. On the 4221HD it was -50dBm.

How do your measured results compare with TVFool Ongley-Rice predictions (after accounting for the antenna gain and cable loss?
The exercise was to objectively measure the difference between the CM4228HD out-of-the-box and the modified version as suggested by Ken Nist's article, so I haven't done any more detailed analysis of the numbers. I gave up on TV-Fool because the chart doesn't show all the actual channel assignments. Also TV-Fool assumes an antenna in open air, mine is indoors.
 
#21 ·
Techluvr,

I tried the mod this week-end and while the result are stronger for most channels, I lost my weak fox on uhf 43. I tried with 2 short 3' RG-6 cables, 2 differents splitters (as combiner at the antenna side) and also removed my attenuator (2 way splitter near the LG), can't catch it.

I don't have any tool to make my cable shorter but I know having good result with those 2 cables.

From your posted measures, can we conclude that the upper uhf range got worse?
 
#23 ·
Just the opposite.:confused: Differences that are +- 1 dB are not significant. With the exception of a 4dB dip at ch 22 and slight dip at ch 34 the numbers for the modified antenna are higher or the same. At ch 60 I measured a 6dB increase and at ch 65 a 9dB increase with the modified antenna.

I use two 6" RG6 cables from the baluns to the combiner. I routed them behind the mast.

I don't know what else to tell you. Antenna design is science, but installing them has an element of magic involved. Every installation will be different. I used this test instrument to optimally aim the antenna and to measure the effect of the modification as it exists in my particular unique installation. Results may vary. Since we don't watch TV on a spectrum analyzer, the final test is whether or not your reception is as good or better with the modified antenna. If there's no difference in the other channels, but you've lost one, then the modification is not right for you.

Also with the CM7777 I found that if I switch to individual UHF/VHF inputs, everything goes haywire. The balancing of the signal with attenuators that I did with the combined input had to be completely re-done when I added the VHF antenna and used the split inputs.
 
#26 ·
Hi Techluvr and all,

Good day to you people,

I am interested in modifying the new CM 4228HD but I have to admit I am not exactly a technically savvy person. I read the instructions posted by Ken but I still have some problems digesting what is written on his page.

Hey Techluvr, care to do me a favor? Can u post more pictures of your modified CM 4228HD and I would try to modify my new antenna according to what is posted on the pictures cos I am not sure what exactly is a phasing harness and how do I replace it with 2 baluns, a combiner plus 2 short cables.

I guess a pic tells a thousand words and I would appreciate if u can help me out with them.

"Fortunately fixing it is trivial: You just replace the phasing harness with two baluns and a combiner. Those plus two short cables will cost you another $25, but a nut-driver is the only tool you need. If the loss in the baluns and combiner is 0.5 dB then the performance will be 0.5 dB less than the B3 plot above"

I hope it is ok with you, thank you so much for that!

regards
sunnydays
 
#27 · (Edited)
I am interested in modifying the new CM 4228HD but I have to admit I am not exactly a technically savvy person. I read the instructions posted by Ken but I still have some problems digesting what is written on his page.
The first thing I want to say is that you should understand what you're trying to achieve before you tackle any project like this. In Mr. Nist's article he predicted an increased gain for the higher number channels by modifying the antenna. In my case, I receive a ch 65 from a broadcast tower about 5Km away, but the reception for some reason was not that great. Modifying the antenna provided a better quality picture for me, and to eliminate any subjective bias on my part, I used the spectrum analyzer and measured a 9dB ( 8 times ) increase in gain at the antenna for that channel. If you have no issues with reception of high channel numbers, I don't know if this modification will do anything for you.

Hey Techluvr, care to do me a favor? Can u post more pictures of your modified CM 4228HD and I would try to modify my new antenna according to what is posted on the pictures cos I am not sure what exactly is a phasing harness and how do I replace it with 2 baluns, a combiner plus 2 short cables.

I guess a pic tells a thousand words and I would appreciate if u can help me out with them.
I did take a few more pictures during the installation...

First you have to remove this thing shown below. There are two plastic caps on each side; pry them off and loosen the hex screws. Remove the rods that lead to the plastic box. There is another plastic do-dad on each side that holds the longer rods in place, just gently pry the rods out of it.



Next, you put a balun on each side. Make sure there's good electrical connection to the vertical rods under the hex screws. The baluns must be hooked up the same way on each side so that they are "in phase". If there's no way to tell which lead is which, you may have to try reversing one if the picture actually gets worse.




Using a splitter in reverse as a combiner, attach 2 equal length short RG-6 cables from the two new baluns to the "outputs" of the splitter. They are really inputs now. Run these cables behind the horizontal rods. The labelled "input" of the splitter is now your output to the TV.

 
#28 ·
I receive a ch 65 from a broadcast tower about 5Km away, but the reception for some reason was not that great. Modifying the antenna provided a better quality picture for me, and to eliminate any subjective bias on my part, I used the spectrum analyzer and measured a 9dB ( 8 times ) gain at the antenna for that channel.
Only 9db ? So my observation/theory about dead spots near strong broadcast towers is correct ?
 
#29 ·
That's a 9dB difference between the original CM4228HD and the modified version. The actual numbers were -57dBm to -48dBm. The station on ch 65 broadcasts 500kW ERP at 215.4 HAAT.

Yes, it is strange that I don't get this station very well. In contrast, ch 14 broadcasts 595kW ERP at 202.3 HAAT, and I measured ~ -33dBm before and after the mod. :confused:

Like you say... black magic, heh ;)
 
#30 ·
That's a 9dB difference between the original CM4228HD and the modified version. The actual numbers were -57dBm to -48dBm. The station on ch 65 broadcasts 500kW ERP at 215.4 HAAT.

Yes, it is strange that I don't get this station very well. In contrast, ch 14 broadcasts 595kW ERP at 202.3 HAAT, and I measured ~ -33dBm before and after the mod.

Like you say... black magic, heh
Perhaps not. It might be that the Mod corrects a SWR problem that occurs at the high end of the band that is not present at the low end of the band.

There is also a possiblity that the CITY-TV-3 ch. 65 antenna has a narrower vertical beam than the OMNI2 ch. 14 antenna. One way to find this out is to see if the original CRTC application documents contain the transmitter outputs, so you can figure out something about the gain of each antenna. It may have an omnidirectional horizontal pattern, but a difference in gains will tell you how sharp the vertical pattern is. (The FCC applications show all this data, so you could figure this out for American stations.)

I'm not sure if TVFool looks at the vertical patterns when calculating ERPs -- I think it migh,t, but of course not for American stations. Also I think the attena pattern info in the American database is only for the horizontal plane. The vertical plane info is used for calculating health and safety limits ( aka Code 6 in Cnaada) for the base of the tower.
 
#31 ·
Perhaps not. It might be that the Mod corrects a SWR problem that occurs at the high end of the band that is not present at the low end of the band.
I'm not actually inferring that there are dark forces at play here. ;) I think Ken Nist's article did mention an impedance mismatch with the balun; that would cause a reflection and a SWR problem, no?

Everything can be explained if we know the physics principles involved. The trick is determining what they are and what corrective action if any is possible. Maybe the attenuation of the higher frequency is greater due to trees and houses in the vicinity. Chain-sawing down the trees and bulldozing the neighbor's house might correct the problem, but those actions could have other non-reception related consequences.

Believe me, I'm not a hack, but sometimes after we have exhausted all the theory, we fall back on the empirical to understand what's happening. If I could find the information you suggest, what more could I do at the antenna to measure and compensate for it?
 
#32 ·
Like you say... black magic, heh
It definately has something to with the Dark Side of the Force, heh.

Everything can be explained if we know the physics principles involved.
Thats the thing, if we know ALL the physics principles involved. Im not convinced anyone can know, nor can know, at least in the near future. There is always something else discovered, heh.
 
#33 ·
If I could find the information you suggest, what more could I do at the antenna to measure and compensate for it?
Nothing really. I think you've taken the correct approach -- try mods that fix any problem channels you have, without causing to lose other channels.

That's why the first question I always ask a poster is "which channels do you want to watch?". For example, for someone who finds little of value on ch. 11 (the only VHF-TV in Ottawa's Rogers Tower at Herberts Corners), then why bother with a a VHF antenna to get a channel you won't watch?

About the closeness to the transmitter -- I'll do some experiments later this week with the interactive TVFool to see if locations close to the CN Tower take the elevation angle into account. (But that wouldn't account for a 10m difference, only for an antenna pattern difference -- hmm, maybe I better pick an American station on a tall tower instead, so we would know that the antenna pattern is accounted for.)
 
#36 ·
All right Ken ! He actually bought one and took it apart.

I was trying to get info on that to do my modeling of it for a few months, but it is near impossible to by asking owners over the internet. Now since hes doing it, I wont have to.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top