Order the meter cct buffer board on line or build one from the diagram. I have the kit but haven’t installed it yet. The meter diodes introduce distortion but the buffer eliminates that. Make sure the screws holding the PCB to the chassis are tight. Good video!
Trevor, this is fun - I feel like I’m looking at my own gear! I have the same generator from another brand and also the IG-18 and had the same result on the THD. Using a Keithley 2015 the Koolertron was about .6xx distortion, almost as high as yours and the IG-18 was less but I don’t remember off the top of my head. I did also replace a bunch of parts in the IG-18 and got the meter kit. Then I spent some time trying to get the voltage up to 43, it was a little low. Never got that exact. But the Heath is fun to use! I’d rather turn a knob than go into menus etc. and it gives a nice solid looking signal. Keep in mind it’s not a precision instrument - it was very good for the price at the time. People go down all kinds of rabbit holes trying to tweak it - not worth it IMO. Just replace some parts and enjoy using it. :)
One trick is that you’ll find that on the IG-18, the signal is a little better when the Fine knob is not at full. At around 3:00 it gives a cleaner signal.
There have been several articles published over the years about how to improve the distortion of this device. IIRC, the meter circuit adds some distortion, which can be reduced with a buffer. Maybe there are diodes across the meter to protect it.
Yes, and I should have done my research before I turned the camera on, lol Lots of viewers emailed me with great suggestions on how to improve the IG-18, as well as the comments here. I'm not done with the IG-18 and when I get more time, I plan to implement some of those suggestions. I also have a new piece of test gear that was delivered the other day which will appear in a future repair video
great content good points, I see you tubers with tons of test gear and wonder how much time they spend to do a proper periodic calibration check. Back in the day (old time saying), probably still today, you have one department devoted to cal lab cause they issued certs for the gear being manufactured everything on the bench had an up to date cal lab sticker. But like they say "good enough for ................... You can go off into infinity getting measurements till you realize the limits of the equipment and call it +/- xxx value .
Thinking back to the days of repair with analog meters and test gear. We still achieved results with a 2% accuracy. Today we don't really need 6 digit resolution to do our daily work. We're getting spoiled
Last time I hooked up my IG-18 the amplitude was bouncing up and down, so now I have a Tremelo. Another person here on TH-cam had the same problem and it was a bad regulator transistor. Changed out the electrolytics, and removed the transistor from the unit, tested it with the Atlas DCA55, and it tested good. Same problem so I put it away and haven't gotten back to it since. I guess the next step would be to check all the resistors since I only spot checked a few and none were that far out. I have been using DSP software on my laptop for distortion. NAK-100 which is made for testing tape decks, but will test signals up to 2 volts due to limitations of the sound card. It will measure down to 0.01 THD and has a spectrum analyzer.
Not shown in this video was some other mods I made which helped a bit more. Some of the changes I did was to use high gain transistors after which sent the generator into a secondary oscillation of about 2Hz that I couldn't stabilize. When I get more time I'd like to revisit the IG-18
Hi first of all thank you very much for your great video and congratulations for your work Then I have a slightly strange question everyone is busting their guts to find a signal generator for example with extremely low distortion But is it really necessary? when you measure a device you basically want it not to take anything away and not to add anything to the input signal you just need to make the difference between input and output obviously taking into account any gain if the spectrum is the same it means that the device under test is transparent i.e. has not added or taken anything out from the input signal, right? So just a decent signal generator could be more than enough
Not a strange question, I also wondered if the distortion meter compares both in- and out of an amp, or that you need the internal low distortion generator in the meter for comparing, or is that only used for calibration of max. deflection. But still, in the first case I think it is slightly more difficult to reproduce (gain) a dirty signal in the meter itself than it is to do with a perfect sine wave. E.g. maybe cross-over distortion can have a fast rise or fall time.
@@erikdenhouter hi thank you for the very kind and valuable advice In the meantime i have thought a little more Actually almost any serious lab at least for audio have very low distortion signal generators that usually are inside a high quality spectrum analyzer The cheap ones can be fine for hobbyist But pros usually want the best they can afford Kind regards gino
@@gino3286 Of course, the wanted standard is as low distortion as possible, the wallet says to buy what you can afford. Spending an extra thousand for something that is really sort of needed once in a blue moon is a waist. You can often find other ways to solve the problem. I can remember working in a repair environment in the end '70 (~15 engineers repairing a few respected brands like Akai and Marantz at the only dealer for the Netherlands), and I can't remember seeing one distortion meter. I think it is more something for measuring equipment than for audio, that is if it is about repair. Design is two steps further.
@@erikdenhouter thanks a lot again Maybe for repair duties a decent scope and a cheap signal generator are more than enough I have seen some videos They just check at what output voltage the 1kHz sine wave top starts to flatten out Very rough measurement Kind regards gino
@@gino3286 "A decent scope and a cheap signal generator", yes, the bare minimum, plus a 2 x 4Ω+4Ω dummy load of 50 or 100 watts, a decent DMM plus a few dirt cheap DMM's, that should get you going.
I built the IG-18 a long time ago. I also modified it with an article I found to lower the THD distortion. I just went and got it out of storage where is has been over the last 20 years. I fired it up and using the QA403 audio analyzer I got 0.063% THD at 1 kHz 0 dBm. Years ago I tagged it with 0.045% THD. The major spectral distortion for the sinewave is at the second and third harmonic and you noted the dominate second harmonic in you video. On a second note: Are you sure your THD analyzer uses a precision internal oscillator to subtract from the test signal to make the THD measurements? I have several analog THD analyzer and none of them use an internal oscillator for nulling. They use a tunable active notch filter to remove or reduce the fundamental signal and then just measure the remaining distortion plus noise signals over a specified bandwidth. I have always been a big fan of Heathkit and thank you for posting this video. Look forward to seeing how low you can get the THD.
Hi Trevor, did you use a linear wall adapter or linear bench powersupply to power that signal generator ? I have the same model of signal generator but different brand, and the 5V switching powersupply that comes with it is very noisy, it also add stray voltage to the negative part of the BNC connectors and this switching powersupply is also underrated. To power my signal generator , i have cut off the cable with the 5.5 x 2.5mm DC jack from a broken notebook charger, add a flat 230V european mains connector to the cut off end, because european mains connectors here in Belgium have also the same distance (19mm) like most bench powersupplies and multimeters use, and power the signal generator with this cable to my linear bench power supply. The output of the signal generator is much cleaner that way, and no stray voltage on the BNC connectors. Grtz
Very good point. I'm using the garbage adapter that came with. I really should eliminate it and get a linear power wart for it. That would also help clean up the bench noise I have. Thanks for the suggestion
Well, if you want comments, I wondered why you kept these carbon resistors and trimmed them with metal film on the bottom. Why not put two metal film as new on the top, and fully get rid of the drift of them ? The distortion directly measured seems on the peaks, not the cross-over, and through the little amp it was also there, but with cross-over, but still measuring almost the same as with direct input. That made me think of the cables as you mentioned too, as if there's a capacitive factor. Maybe try the cables as short as possible, say 20 cm to see if that is the right way to think ? I als think about impedance mismatching, but I don't know if that kan make such differences, I lack experience there.
All good points. I should have yanked all the carbon comp resistors and replaced them, it's not too late to make that happen. I don't see temp drift a big problem. The IG-18 uses about 5 watts to run and internal heating is minimum
revor, I have a HP 8640B with no RF output. The Power meter light does not turn on. The unit turns on but no RF output and no counter output movement when the internal frequency dial is turned. The power supply LEDs are one. Can you please provide suggestions. Al
I'm surprised you kept some carbon composite resistors in there Trevor, metal film generally introduce less low frequency thermal noise and are a lot more temp' stable.
Nice work on the operation and repairs you did. Your explanation of the operation of the distortion analyzer differs from the instruction book. This THD+N meter and all the ones I have owned use a sharp notch filter that filters out the fundamental and meters what is left over. The weighting button puts in a set of filters to measure noise on telephone lines, it is of no use in amplifier measurements that I can see.
All really interesting. And this ⒶⓃⒶⓁⓄⒼ ⓂⒶⒼⒾⒸ is just that, magical "to me". I've been subbed but haven't see this channel pop up in the ones pushed by YT since I don't know when. Once again I'm convinced that TH-cam has NO IDEA what I like. Gosh forbid if you click on (1) goofy video you immediately get inundated with them. Like brushing up against position ivy or a leaky pen in the hand. Of course the answer is to "ring your bell"... but I'm hesitant because of the way that sounds But [it] been' done! Till the next video. Cheers from So.Ca.USA 3rd house on the left. p.s. Love the F242. But from your description of it's operation, if I was able to get one, would have to WORK out of the BOX
Like I said in the vid, the AWA is great for setting up FM tuners for low distortion. I'm quickly finding out this meter is more difficult to measure amplifiers. There are better suited test gear for measuring amplifier distortion such as HP 8903 and others. I have videos on the AWA in my back catalog if you're interested in seeing me get antiquated with it
Thanks for the quality content, Trevor. Always much appreciated and enjoyable viewing.👍
Order the meter cct buffer board on line or build one from the diagram. I have the kit but haven’t installed it yet. The meter diodes introduce distortion but the buffer eliminates that. Make sure the screws holding the PCB to the chassis are tight. Good video!
Trevor, this is fun - I feel like I’m looking at my own gear! I have the same generator from another brand and also the IG-18 and had the same result on the THD. Using a Keithley 2015 the Koolertron was about .6xx distortion, almost as high as yours and the IG-18 was less but I don’t remember off the top of my head. I did also replace a bunch of parts in the IG-18 and got the meter kit. Then I spent some time trying to get the voltage up to 43, it was a little low. Never got that exact. But the Heath is fun to use! I’d rather turn a knob than go into menus etc. and it gives a nice solid looking signal. Keep in mind it’s not a precision instrument - it was very good for the price at the time. People go down all kinds of rabbit holes trying to tweak it - not worth it IMO. Just replace some parts and enjoy using it. :)
One trick is that you’ll find that on the IG-18, the signal is a little better when the Fine knob is not at full. At around 3:00 it gives a cleaner signal.
There have been several articles published over the years about how to improve the distortion of this device. IIRC, the meter circuit adds some distortion, which can be reduced with a buffer. Maybe there are diodes across the meter to protect it.
Yes, and I should have done my research before I turned the camera on, lol
Lots of viewers emailed me with great suggestions on how to improve the IG-18, as well as the comments here.
I'm not done with the IG-18 and when I get more time, I plan to implement some of those suggestions.
I also have a new piece of test gear that was delivered the other day which will appear in a future repair video
great content good points, I see you tubers with tons of test gear and wonder how much time they spend to do a proper periodic calibration check. Back in the day (old time saying), probably still today, you have one department devoted to cal lab cause they issued certs for the gear being manufactured everything on the bench had an up to date cal lab sticker. But like they say "good enough for ................... You can go off into infinity getting measurements till you realize the limits of the equipment and call it +/- xxx value .
Thinking back to the days of repair with analog meters and test gear. We still achieved results with a 2% accuracy. Today we don't really need 6 digit resolution to do our daily work. We're getting spoiled
@@TrevorsBench true, 6 digit resolution only to those who are willing to pay for it. vinyl, tape,is my ear, over CD, digital
Last time I hooked up my IG-18 the amplitude was bouncing up and down, so now I have a Tremelo. Another person here on TH-cam had the same problem and it was a bad regulator transistor. Changed out the electrolytics, and removed the transistor from the unit, tested it with the Atlas DCA55, and it tested good. Same problem so I put it away and haven't gotten back to it since. I guess the next step would be to check all the resistors since I only spot checked a few and none were that far out.
I have been using DSP software on my laptop for distortion. NAK-100 which is made for testing tape decks, but will test signals up to 2 volts due to limitations of the sound card. It will measure down to 0.01 THD and has a spectrum analyzer.
Not shown in this video was some other mods I made which helped a bit more.
Some of the changes I did was to use high gain transistors after which sent the generator into a secondary oscillation of about 2Hz that I couldn't stabilize. When I get more time I'd like to revisit the IG-18
Pour aller plus loin, il faut un QA403 avec ses 0.0001% de distorsion 👍🛸
Hi first of all thank you very much for your great video and congratulations for your work
Then I have a slightly strange question
everyone is busting their guts to find a signal generator for example with extremely low distortion
But is it really necessary?
when you measure a device you basically want it not to take anything away and not to add anything to the input signal
you just need to make the difference between input and output obviously taking into account any gain
if the spectrum is the same it means that the device under test is transparent i.e. has not added or taken anything out from the input signal, right?
So just a decent signal generator could be more than enough
Not a strange question, I also wondered if the distortion meter compares both in- and out of an amp, or that you need the internal low distortion generator in the meter for comparing, or is that only used for calibration of max. deflection. But still, in the first case I think it is slightly more difficult to reproduce (gain) a dirty signal in the meter itself than it is to do with a perfect sine wave. E.g. maybe cross-over distortion can have a fast rise or fall time.
@@erikdenhouter hi thank you for the very kind and valuable advice
In the meantime i have thought a little more
Actually almost any serious lab at least for audio have very low distortion signal generators that usually are inside a high quality spectrum analyzer
The cheap ones can be fine for hobbyist
But pros usually want the best they can afford Kind regards gino
@@gino3286 Of course, the wanted standard is as low distortion as possible, the wallet says to buy what you can afford. Spending an extra thousand for something that is really sort of needed once in a blue moon is a waist. You can often find other ways to solve the problem.
I can remember working in a repair environment in the end '70 (~15 engineers repairing a few respected brands like Akai and Marantz at the only dealer for the Netherlands), and I can't remember seeing one distortion meter. I think it is more something for measuring equipment than for audio, that is if it is about repair. Design is two steps further.
@@erikdenhouter thanks a lot again
Maybe for repair duties a decent scope and a cheap signal generator are more than enough
I have seen some videos
They just check at what output voltage the 1kHz sine wave top starts to flatten out
Very rough measurement
Kind regards gino
@@gino3286 "A decent scope and a cheap signal generator", yes, the bare minimum, plus a 2 x 4Ω+4Ω dummy load of 50 or 100 watts, a decent DMM plus a few dirt cheap DMM's, that should get you going.
The meter load affects the distortion. Look for an old article called Greening the IG-18.
I built the IG-18 a long time ago. I also modified it with an article I found to lower the THD distortion. I just went and got it out of storage where is has been over the last 20 years. I fired it up and using the QA403 audio analyzer I got 0.063% THD at 1 kHz 0 dBm. Years ago I tagged it with 0.045% THD. The major spectral distortion for the sinewave is at the second and third harmonic and you noted the dominate second harmonic in you video. On a second note: Are you sure your THD analyzer uses a precision internal oscillator to subtract from the test signal to make the THD measurements? I have several analog THD analyzer and none of them use an internal oscillator for nulling. They use a tunable active notch filter to remove or reduce the fundamental signal and then just measure the remaining distortion plus noise signals over a specified bandwidth. I have always been a big fan of Heathkit and thank you for posting this video. Look forward to seeing how low you can get the THD.
Hi Trevor, did you use a linear wall adapter or linear bench powersupply to power that signal generator ?
I have the same model of signal generator but different brand, and the 5V switching powersupply that comes with it is very noisy, it also add stray voltage to the negative part of the BNC connectors and this switching powersupply is also underrated. To power my signal generator , i have cut off the cable with the 5.5 x 2.5mm DC jack from a broken notebook charger, add a flat 230V european mains connector to the cut off end, because european mains connectors here in Belgium have also the same distance (19mm) like most bench powersupplies and multimeters use, and power the signal generator with this cable to my linear bench power supply. The output of the signal generator is much cleaner that way, and no stray voltage on the BNC connectors. Grtz
Very good point. I'm using the garbage adapter that came with. I really should eliminate it and get a linear power wart for it. That would also help clean up the bench noise I have. Thanks for the suggestion
@@TrevorsBench a idea for a upcoming video maybe ? Grtz
Well, if you want comments, I wondered why you kept these carbon resistors and trimmed them with metal film on the bottom. Why not put two metal film as new on the top, and fully get rid of the drift of them ?
The distortion directly measured seems on the peaks, not the cross-over, and through the little amp it was also there, but with cross-over, but still measuring almost the same as with direct input. That made me think of the cables as you mentioned too, as if there's a capacitive factor. Maybe try the cables as short as possible, say 20 cm to see if that is the right way to think ? I als think about impedance mismatching, but I don't know if that kan make such differences, I lack experience there.
All good points. I should have yanked all the carbon comp resistors and replaced them, it's not too late to make that happen. I don't see temp drift a big problem. The IG-18 uses about 5 watts to run and internal heating is minimum
revor, I have a HP 8640B with no RF output. The Power meter light does not turn on. The unit turns on but no RF output and no counter output movement when the internal frequency dial is turned. The power supply LEDs are one. Can you please provide suggestions. Al
I'm surprised you kept some carbon composite resistors in there Trevor, metal film generally introduce less low frequency thermal noise and are a lot more temp' stable.
amalgamated wireless australia made test gear?
Put a regulator IC in the Power Supply?
The supply is already regulated to +43v
Did you change all the capacitors ?
Yes
Nice work on the operation and repairs you did. Your explanation of the operation of the distortion analyzer differs from the instruction book. This THD+N meter and all the ones I have owned use a sharp notch filter that filters out the fundamental and meters what is left over. The weighting button puts in a set of filters to measure noise on telephone lines, it is of no use in amplifier measurements that I can see.
All really interesting. And this ⒶⓃⒶⓁⓄⒼ ⓂⒶⒼⒾⒸ is just that, magical "to me". I've been subbed but haven't see this channel pop up in the ones pushed by YT since I don't know when. Once again I'm convinced that TH-cam has NO IDEA what I like. Gosh forbid if you click on (1) goofy video you immediately get inundated with them. Like brushing up against position ivy or a leaky pen in the hand. Of course the answer is to "ring your bell"... but I'm hesitant because of the way that sounds But [it] been' done! Till the next video. Cheers from So.Ca.USA 3rd house on the left. p.s. Love the F242. But from your description of it's operation, if I was able to get one, would have to WORK out of the BOX
Like I said in the vid, the AWA is great for setting up FM tuners for low distortion.
I'm quickly finding out this meter is more difficult to measure amplifiers.
There are better suited test gear for measuring amplifier distortion such as HP 8903 and others.
I have videos on the AWA in my back catalog if you're interested in seeing me get antiquated with it