TH3JR Analysis
Created by on 2019-12-07Written by Jeff King, ZL4AI
Purpose of this Article:
This Part 1 is the first of a series of articles showing analysis of the Hygain TH3JR showing initial gain on 20 metres is only -13.9 dBi. Subsequent articles show how the NEC model was further developed and the antenna altered to provide a theoretical gain of 11.8 dBi on 20 metres. The purpose of the article is to request feedback from amateurs who have used and modified the TH3JR to gain their knowledge of how well the antenna worked installed to the factory manual and any experiences they may have of modification and performance improvement.
The next version of this post will show how the antenna was modified.
What Is the Real Performance?
I have had a TH3JR for some time. I had not got to putting it up because I was unconvinced by the Hygain Specification which says:
The questions I had were:
What does up to 8 dBbi mean? One would assume obviously this would be a simple statement for the best band so that was likely to be the performance on 10 meters. So, the question I had was "what was the gain on 20 meters and 15 meters?"
10 meters is never open, so it is of little interest to a DXer.
Really the performance on 20 meters is really what matters. so, what was the gain 20 metres?
Well this should be easy to answer in theory. All I need is the NEC model and I can run this through 4NEC.
I have searched the internet for years, but could I find an NEC model? No. In fact here are very few NEC models or commercially made beams form the per NEC era.
So eventually in 2019 I decided to dedicate considerable time (4 weeks) to creating an NEC model and finding out what the beam really did put out.
Building The Model
Finding in the dimensions from the HighGain was little challenging because the copies on the internet are illegible. Obviously a quality control issue with MFJ who put out the current manuals.
However eventually by recalculating the imperial feet and inches I was able to get all the dimensions.
Traps
Information on the TRAPS was nonexistent except for some values of dip frequencies using a dip meter.
I was able to measure length form the traps I held
To find resonant frequencies I connected a Rig Expert AA-54 antenna analyzer across the trap with short lengths of insulated wire. Traps have to be kept at least 400 mm away from other metal objects to get a repeatable reading. To identify the resonant frequency scan from 0 to 30MHZ and using the R+Ji series or parallel view indentify the region where inductance changes to capacitance. This is shown in figure 4.67 from the 2007 ARRL Handbook. The theory of on resonance is explained in detail in Chapter 4 and so if you want to understand the basis of these calculations read Chapter 4 on L / C parallel circuits at resonance.
The measured values of resonant frequency are very different to the trap dip figures Hygain published in the 1970s.
To measure the inductance of coils I took the outer capacitor tube off the trap and measured the coil inductance in the region where the value was low (say between 0.08 to 2.0 as shown in the above figure 4.67) well before the inductance values rapidly increase at resonance. The rapidly increasing values are not typical of trap operation and should be ignored.
Calculate Capacitance
C = 25,330 / ((F^2) . L)
Alternatively, you can measure trap capacitance using an oscilloscope to establish the time for charging to 5/8 height of square pulse wave. Many examples of how to measure this way are available on Youtube.com.
Calculate Performance Through Several Steps
1. Stage 1: Analysis of the original Hygain TH3JR complete antenna model
The TH3JR is a complex electrical structure: where the boom is electrically connected to the Director and Reflector: The matching is via a hairpin which extends towards the reflector and connects electrically in the centre to the boom at point Hygain say is neutral.
The Model analyzed is below.
On 14.225 MHz showed a very unimpressive -13.9 dBi off the back of the beam and current was formed on only half the elements: Essentially a very low performance of rotatable dipole.
Look above at those currents in red. Distorted due to the input matching system of due to a hairpin attached to the boom.
This matching causes:
currents along currents along the boom. process and
the current drops significantly on the right hand side Driven Element (when viewed indirection Reflector to Director ).
On 21.225 MHz showed a very unimpressive -1.3Bi off the back of the beam
on 15 MHZ had currents formed only partly across the DE DIR and REF elements to 10m traps: Is better currents but these should extend at least out to the 15M trap.
On 28.050 MHz showed a very impressive 13.46 dBi off the front of the beam
10 MHZ had high currents formed across the 10m sections of elements: shown above and below.
Conclusion: This antenna is of little or no use to me as it has no gain or performance on 20 meters.
Next Part of this article will tell you how the antenna model was analyzed and altered in the model to raise the 20m dBi peak off the front of the beam to 11.8dBi. Plots of the 10m and 20m band will also be supplied.
NEC File for TH3JR analyzed in Part 1.
CM Multi Band TH3JR (20m,15m,10m)
CM This is the full antenna with actual trap measurements
CM Lengths optimized for 20 meter above average ground
CE
SY delen = 0.737 'Director end length
SY DIRfifc = 2.46pF '15m trap capacitance DIR&REF Actual measure 2.46pF
SY DIRfifi = 47.87uH '15m trap inductance DIR&REF Actual measure 47.87uH
SY DEfii=46.08uH '15MTRAP DE inductance Actual measure = 46.08uH
SY DEfic=2.34pF '15MTRAP DE capacitance Actual measure =2.34pF
SY DEtei=7.56uH '10MTRAP DE inductance Actual measure =7.56uH
SY DEtec=10.32pF '10MTRAP DE capacitance Actual measure =10.32pF
SY REFteni= 11.09uH '10M TRAP REF Actual measure = 11.09uH
SY REFtenc= 9.00pF '10M TRAP REF Actual measure = 9.00 pF
SY DIRteni=39.243uH '10M TRAP DIR Actual measure =39.24uH
SY DIRtenc=2.11pF '10M TRAP DIR Actual measure =2.11pF
SY REFfifc = 2.46pF '15m trap capacitance DIR&REF Actual measure 2.46pF
SY REFfifi = 47.87uH '15m trap inductance DIR&REF Actual measure 47.87uH
CM CM DRIVEN elementGW3600.1252001.845202.20E-02GW460-0.125200-1.845202.20E-02GW5301.8452002.131201.60E-02GW630-1.845200-2.131201.60E-02GW11302.1312002.366201.60E-02GW1230-2.131200-2.366201.60E-02GW7202.3662002.633201.90E-02GW820-2.366200-2.633201.90E-02GW21202.6332002.892201.90E-02GW3120-2.633200-2.892201.90E-02GW9802.8922003.756201.10E-02GW1080-2.892200-3.756201.10E-02CM REFLECTOR GW5111.9050.14201.9050202.20E-02GW5911.9050201.905-0.14202.20E-02GW5381.9050.14201.9051.845202.20E-02GW5481.905-0.14201.905-1.845202.20E-02GW5521.9051.845201.9052.137201.60E-02GW5621.905-1.845201.905-2.137201.60E-02GW6121.9052.137201.9052.333201.60E-02GW6221.905-2.137201.905-2.333201.60E-02GW5711.9052.333201.9052.492201.90E-02GW5811.905-2.333201.905-2.492201.90E-02GW7121.9052.492201.9052.729201.90E-02GW8121.905-2.492201.905-2.729201.90E-02GW59121.9052.729201.9053.481201.10E-02GW60121.905-2.729201.905-3.481201.10E-02CM DIRECTOR element GW1011-1.6540.1420-1.6540202.20E-02GW1021-1.654020-1.654-0.14202.20E-02GW1038-1.6540.1420-1.6541.845202.20E-02GW1048-1.654-0.1420-1.654-1.845202.20E-02GW1052-1.6541.84520-1.6542.594201.60E-02GW1062-1.654-1.84520-1.654-2.594201.60E-02GW1112-1.6542.59420-1.6542.791201.60E-02GW1122-1.654-2.59420-1.654-2.791201.60E-02GW1072-1.6542.79120-1.6543.070201.90E-02GW1082-1.654-2.79120-1.654-3.070201.90E-02GW1212-1.6543.07020-1.6543.307201.90E-02GW1312-1.654-3.07020-1.654-3.307201.90E-02GW10912-1.6543.30720-1.6544.082201.10E-02GW11012-1.654-3.30720-1.654-4.082201.10E-02cMCM HAIR PIN includedGW204100.12520-0.0550.12520.0460.008GW20510-0.12520-0.055-0.12520.0460.008GW2061-0.0550.12520.046-0.1500.02220.0460.005GW2071-0.055-0.12520.046-0.150-0.02220.0460.005GW20810-0.8660.02220.046-0.1500.02220.0460.005GW2091-0.8660.02220.046-0.866020.0460.003GW2101-0.866020-0.866020.0460.008GW2111-0.866020.046-0.866-0.02220.0460.003GW21210-0.866-0.02220.046-0.150-0.02220.0460.005CM BOOMX1Y1Z1X2Y2Z2GW20161.90502000203.20E-02GW2026-1.654020-0.8660203.20E-02GW2036-0.86602000203.20E-02GE-1LD61111100DEteiDEtecLD61211100DEteiDEtecLD62111100DEfiiDEficLD63111100DEfiiDEficLD66111100DIRteniDIRtencLD66211100DIRteniDIRtencLD67111100DIRfifiDIRfifcLD68111100DIRfifiDIRfifcLD611111100REFteniREFtencLD611211100REFteniREFtencLD612111100REFfifiREFfifcLD613111100REFfifiREFfifcLD530024900000'Alum. 6061-T6LD540024900000'Alum. 6061-T6LD550024900000'Alum. 6061-T6LD560024900000'Alum. 6061-T6LD5110024900000'Alum. 6061-T6LD5120024900000'Alum. 6061-T6LD570024900000'Alum. 6061-T6LD580024900000'Alum. 6061-T6LD5210024900000'Alum. 6061-T6LD5310024900000'Alum. 6061-T6LD590024900000'Alum. 6061-T6LD5100024900000'Alum. 6061-T6LD5510024900000'Alum. 6061-T6LD5530024900000'Alum. 6061-T6LD5540024900000'Alum. 6061-T6LD5550024900000'Alum. 6061-T6LD5560024900000'Alum. 6061-T6LD5610024900000'Alum. 6061-T6LD5620024900000'Alum. 6061-T6LD5570024900000'Alum. 6061-T6LD5580024900000'Alum. 6061-T6LD5710024900000'Alum. 6061-T6LD5810024900000'Alum. 6061-T6LD5590024900000'Alum. 6061-T6LD5600024900000LD51010024900000'Alum. 6061-T6LD51030024900000'Alum. 6061-T6LD51040024900000'Alum. 6061-T6LD51050024900000'Alum. 6061-T6LD51060024900000'Alum. 6061-T6LD51110024900000'Alum. 6061-T6LD51120024900000'Alum. 6061-T6LD51070024900000'Alum. 6061-T6LD51080024900000'Alum. 6061-T6LD51210024900000'Alum. 6061-T6LD51310024900000'Alum. 6061-T6LD51090024900000'Alum. 6061-T6LD51100024900000LD52010024900000'Alum. 6061-T6LD52020024900000'Alum. 6061-T6LD52030024900000'Alum. 6061-T6LD52040024900000'Alum. 6061-T6LD52050024900000'Alum. 6061-T6LD52060024900000'Alum. 6061-T6LD52070024900000'Alum. 6061-T6LD52080024900000'Alum. 6061-T6LD52090024900000'Alum. 6061-T6LD52100024900000'Alum. 6061-T6LD52110024900000'Alum. 6061-T6LD52120024900000'Alum. 6061-T6GN2000130.005EKEX0310100FR000014.2250EN
Copyright of everything in this article belongs to Jeff King
| ZL4AI | 2021-05-01 | |
|---|---|---|
| TH3JR Analysis | ||
| NEC File for TH3JR analyzed in Part 1. CM Multi Band TH3JR (20m,15m,10m) CM This is the full antenna with actual trap measurements CM Lengths optimized for 20 meter above average ground CE SY delen = 0.737 'Director end length SY DIRfifc = 2.46pF '15m trap capacitance DIR&REF Actual measure 2.46pF SY DIRfifi = 47.87uH '15m trap inductance DIR&REF Actual measure 47.87uH SY DEfii=46.08uH '15MTRAP DE inductance Actual measure = 46.08uH SY DEfic=2.34pF '15MTRAP DE capacitance Actual measure =2.34pF SY DEtei=7.56uH '10MTRAP DE inductance Actual measure =7.56uH SY DEtec=10.32pF '10MTRAP DE capacitance Actual measure =10.32pF SY REFteni= 11.09uH '10M TRAP REF Actual measure = 11.09uH SY REFtenc= 9.00pF '10M TRAP REF Actual measure = 9.00 pF SY DIRteni=39.243uH '10M TRAP DIR Acutal measure =39.24uH SY DIRtenc=2.11pF '10M TRAP DIR Acutal measure =2.11pF SY REFfifc = 2.46pF '15m trap capacitance DIR&REF Actual measure 2.46pF SY REFfifi = 47.87uH '15m trap inductance DIR&REF Actual measure 47.87uH CM CM DRIVEN element GW 3 6 0 0.125 20 0 1.845 20 2.20E-02 GW 4 6 0 -0.125 20 0 -1.845 20 2.20E-02 GW 5 3 0 1.845 20 0 2.131 20 1.60E-02 GW 6 3 0 -1.845 20 0 -2.131 20 1.60E-02 GW 11 3 0 2.131 20 0 2.366 20 1.60E-02 GW 12 3 0 -2.131 20 0 -2.366 20 1.60E-02 GW 7 2 0 2.366 20 0 2.633 20 1.90E-02 GW 8 2 0 -2.366 20 0 -2.633 20 1.90E-02 GW 21 2 0 2.633 20 0 2.892 20 1.90E-02 GW 31 2 0 -2.633 20 0 -2.892 20 1.90E-02 GW 9 8 0 2.892 20 0 3.756 20 1.10E-02 GW 10 8 0 -2.892 20 0 -3.756 20 1.10E-02 CM REFLECTOR GW 51 1 1.905 0.14 20 1.905 0 20 2.20E-02 GW 59 1 1.905 0 20 1.905 -0.14 20 2.20E-02 GW 53 8 1.905 0.14 20 1.905 1.845 20 2.20E-02 GW 54 8 1.905 -0.14 20 1.905 -1.845 20 2.20E-02 GW 55 2 1.905 1.845 20 1.905 2.137 20 1.60E-02 GW 56 2 1.905 -1.845 20 1.905 -2.137 20 1.60E-02 GW 61 2 1.905 2.137 20 1.905 2.333 20 1.60E-02 GW 62 2 1.905 -2.137 20 1.905 -2.333 20 1.60E-02 GW 57 1 1.905 2.333 20 1.905 2.492 20 1.90E-02 GW 58 1 1.905 -2.333 20 1.905 -2.492 20 1.90E-02 GW 71 2 1.905 2.492 20 1.905 2.729 20 1.90E-02 GW 81 2 1.905 -2.492 20 1.905 -2.729 20 1.90E-02 GW 59 12 1.905 2.729 20 1.905 3.481 20 1.10E-02 GW 60 12 1.905 -2.729 20 1.905 -3.481 20 1.10E-02 CM DIRECTOR element GW 101 1 -1.654 0.14 20 -1.654 0 20 2.20E-02 GW 102 1 -1.654 0 20 -1.654 -0.14 20 2.20E-02 GW 103 8 -1.654 0.14 20 -1.654 1.845 20 2.20E-02 GW 104 8 -1.654 -0.14 20 -1.654 -1.845 20 2.20E-02 GW 105 2 -1.654 1.845 20 -1.654 2.594 20 1.60E-02 GW 106 2 -1.654 -1.845 20 -1.654 -2.594 20 1.60E-02 GW 111 2 -1.654 2.594 20 -1.654 2.791 20 1.60E-02 GW 112 2 -1.654 -2.594 20 -1.654 -2.791 20 1.60E-02 GW 107 2 -1.654 2.791 20 -1.654 3.070 20 1.90E-02 GW 108 2 -1.654 -2.791 20 -1.654 -3.070 20 1.90E-02 GW 121 2 -1.654 3.070 20 -1.654 3.307 20 1.90E-02 GW 131 2 -1.654 -3.070 20 -1.654 -3.307 20 1.90E-02 GW 109 12 -1.654 3.307 20 -1.654 4.082 20 1.10E-02 GW 110 12 -1.654 -3.307 20 -1.654 -4.082 20 1.10E-02 cM CM HAIR PIN included GW 204 1 0 0.125 20 -0.055 0.125 20.046 0.008 GW 205 1 0 -0.125 20 -0.055 -0.125 20.046 0.008 GW 206 1 -0.055 0.125 20.046 -0.150 0.022 20.046 0.005 GW 207 1 -0.055 -0.125 20.046 -0.150 -0.022 20.046 0.005 GW 208 10 -0.866 0.022 20.046 -0.150 0.022 20.046 0.005 GW 209 1 -0.866 0.022 20.046 -0.866 0 20.046 0.003 GW 210 1 -0.866 0 20 -0.866 0 20.046 0.008 GW 211 1 -0.866 0 20.046 -0.866 -0.022 20.046 0.003 GW 212 10 -0.866 -0.022 20.046 -0.150 -0.022 20.046 0.005 CM BOOM X1 Y1 Z1 X2 Y2 Z2 GW 201 6 1.905 0 20 0 0 20 3.20E-02 GW 202 6 -1.654 0 20 -0.866 0 20 3.20E-02 GW 203 6 -0.866 0 20 0 0 20 3.20E-02 GE -1 LD 6 11 1 1 100 DEtei DEtec LD 6 12 1 1 100 DEtei DEtec LD 6 21 1 1 100 DEfii DEfic LD 6 31 1 1 100 DEfii DEfic LD 6 61 1 1 100 DIRteni DIRtenc LD 6 62 1 1 100 DIRteni DIRtenc LD 6 71 1 1 100 DIRfifi DIRfifc LD 6 81 1 1 100 DIRfifi DIRfifc LD 6 111 1 1 100 REFteni REFtencLD 6 112 1 1 100 REFteni REFtenc LD 6 121 1 1 100 REFfifi REFfifc LD 6 131 1 1 100 REFfifi REFfifc LD 5 3 0 0 24900000 'Alum. 6061-T6 LD 5 4 0 0 24900000 'Alum. 6061-T6 LD 5 5 0 0 24900000 'Alum. 6061-T6 LD 5 6 0 0 24900000 'Alum. 6061-T6 LD 5 11 0 0 24900000 'Alum. 6061-T6 LD 5 12 0 0 24900000 'Alum. 6061-T6 LD 5 7 0 0 24900000 'Alum. 6061-T6 LD 5 8 0 0 24900000 'Alum. 6061-T6 LD 5 21 0 0 24900000 'Alum. 6061-T6 LD 5 31 0 0 24900000 'Alum. 6061-T6 LD 5 9 0 0 24900000 'Alum. 6061-T6 LD 5 10 0 0 24900000 'Alum. 6061-T6 LD 5 51 0 0 24900000 'Alum. 6061-T6 LD 5 53 0 0 24900000 'Alum. 6061-T6 LD 5 54 0 0 24900000 'Alum. 6061-T6 LD 5 55 0 0 24900000 'Alum. 6061-T6 LD 5 56 0 0 24900000 'Alum. 6061-T6 LD 5 61 0 0 24900000 'Alum. 6061-T6 LD 5 62 0 0 24900000 'Alum. 6061-T6 LD 5 57 0 0 24900000 'Alum. 6061-T6 LD 5 58 0 0 24900000 'Alum. 6061-T6 LD 5 71 0 0 24900000 'Alum. 6061-T6 LD 5 81 0 0 24900000 'Alum. 6061-T6 LD 5 59 0 0 24900000 'Alum. 6061-T6 LD 5 60 0 0 24900000 LD 5 101 0 0 24900000 'Alum. 6061-T6 LD 5 103 0 0 24900000 'Alum. 6061-T6 LD 5 104 0 0 24900000 'Alum. 6061-T6 LD 5 105 0 0 24900000 'Alum. 6061-T6 LD 5 106 0 0 24900000 'Alum. 6061-T6 LD 5 111 0 0 24900000 'Alum. 6061-T6 LD 5 112 0 0 24900000 'Alum. 6061-T6 LD 5 107 0 0 24900000 'Alum. 6061-T6 LD 5 108 0 0 24900000 'Alum. 6061-T6 LD 5 121 0 0 24900000 'Alum. 6061-T6 LD 5 131 0 0 24900000 'Alum. 6061-T6 LD 5 109 0 0 24900000 'Alum. 6061-T6 LD 5 110 0 0 24900000 LD 5 201 0 0 24900000 'Alum. 6061-T6 LD 5 202 0 0 24900000 'Alum. 6061-T6 LD 5 203 0 0 24900000 'Alum. 6061-T6 LD 5 204 0 0 24900000 'Alum. 6061-T6 LD 5 205 0 0 24900000 'Alum. 6061-T6 LD 5 206 0 0 24900000 'Alum. 6061-T6 LD 5 207 0 0 24900000 'Alum. 6061-T6 LD 5 208 0 0 24900000 'Alum. 6061-T6 LD 5 209 0 0 24900000 'Alum. 6061-T6 LD 5 210 0 0 24900000 'Alum. 6061-T6 LD 5 211 0 0 24900000 'Alum. 6061-T6 LD 5 212 0 0 24900000 'Alum. 6061-T6 GN 2 0 0 0 13 0.005 EK EX 0 3 1 0 1 0 0 FR 0 0 0 0 14.225 0 EN Copyright of this NEC model belongs to Jeff King | ||
| ZL4AI | 2021-05-01 | |
|---|---|---|
| Re: TH3JR Analysis | ||
| For some reason EHam did not publish the full model I provided. Eham only published part of (say 15%) of the NEC model. The NEC model is about about 3 A4 sized pages and contains data on all the traps. You can NOT run the model from the little information now available on Eham. For any person wanting the entire model which is about 3 A4 sized pages this can be obtained via my QRZ page https://www.qrz.com/db/zl4ai. There you will find a link to the ZL4AI Web page which has a section with the full analysis of TH3JR. This page also explains I have not written part 2. The reason is I have not written is; I spent many hours (40+) optimizing the TH3JR by changing the inductance and capacitance of the traps. This provided really good gain in the region of 11.x dBi on the 20 meter band which was what I was interested in. Gain was good on 15m and 10 m too. So I should have written part 2, but... I was frustrated that the traps lost about 10% or more of the power, and to make adjustments there was a lot of work changing the coils in the traps and capacitance. [As a professional engineer power loss frustrates me if the design can be changed to gain another say 15% output] My reason for the modeling the TH3JR initially was to find out what performance was on 20m because the information from Hygain was so vague I could not be sure what was really the expected gain on 20m. The outcome was far far worse than I ever imagined. The advantage of 4NEC is it really enables you to Compare like with like as long as you put time in. So I then turned to Mini Horse (MH)-antenna. http://www.sm0dtk.se/antennas.htm I built the model and optimized that for 11.5dB at 17 degrees This was a much better solution which gave a smaller light weight yagi on 20m. All I had to do was adjust wire lengths of wires. Much easier and much less work to make Mini Horse than alter Th4JR traps. I am less interested in 10m and 15m because those bands are never open especially during this period of the low in the 11 years cycle. I have not built the Mini Horse yet but will do so when I get time and 20m propagation improves. Reply to a comment by : AF5CC on 2019-12-25 On 20m and 15m you measured the gain of of the back of the beam. Then on 10m you measured the gain in the forward direction? Why this change? | ||
| KK6UE | 2020-08-23 | |
|---|---|---|
| TH3JR Analysis | ||
| Hi Jeff: I also have on Th3JR and notice that it seems a little lackluster on the forward gain on 20 meters. Could this be caused by the short boom. Have you written a part 2 yet or are you in the process of doing so? Let me know Thanks Bob KK6UE | ||
| ZL4AI | 2019-12-29 | |
|---|---|---|
| Re: TH3JR Analysis | ||
| Hi CQ W5TD, Thanks for your observation. I did not change the way the NEC model showed gain. It really surprised me too that the model of the antenna in standard form had more gain came off the back of the beam on 15m and 20m. Copy the NEC model and try running it. You will see how the model shows the results. I am interested in any feed back on how well the TH3JR actually worked on 20 m. When I publish the modified version with revised traps you will see that the gain can be increased on 20m to 11.8 dBi in the forward diorection. 73s Reply to a comment by : AF5CC on 2019-12-25 On 20m and 15m you measured the gain of of the back of the beam. Then on 10m you measured the gain in the forward direction? Why this change? | ||
| AF5CC | 2019-12-25 | |
|---|---|---|
| TH3JR Analysis | ||
| On 20m and 15m you measured the gain of of the back of the beam. Then on 10m you measured the gain in the forward direction? Why this change? | ||
