How to avoid buying a  $1000 HP 8640  SIGNAL GENERATOR (SIG-GEN) and learn to use DDS and your scope to do receiver alignment.
This is a fun topic, it is about how to use modern electronic devices that didn't exist before. (for sure not in 1975)
I got all this to work great, the DDS and all parts.

A Preamble:
Or as they say , "How to cut a fat hog !"(skin flints unite)
As it applies to doing my Heath radio alignments. (SB104 tops my list)

The best web page covering advance swept alignments are here.(for the fully equipped shop/lab/shack)
First some odd facts:
There are 3 alignment books, non-A, model A and the service bulletin #11 (I'll call it SB11) from (H3LL). The SB11 is not a book its a sheet, and today is a text file.(I made it a PDF So I can add comments easy)
The SB11 tells you you need a $10,000 lab of listed equipment , this is  100% bogus.
 They really ment to say , "only to do all service and repairs", and for some alignment (if you desire better out comes, page 245 in the Heath manual)
The fact is, the first book shows ZERO tools,needed, if you read it, it shows, using only the radio's built in features, (the panel meter and 2 hidden test TP jumpers) In fact, this was a selling feature then!
The SB11 even calls out a  specific  signal generator not sold in the USA. (with unknown specs, and hard to track , or discover, obsure is the word here)
It is true as the 2 books state a better alignment, can be attained for the "G" BPF card, if you use a real  SIG GEN. (it's no lie that)
Not only that, but it's only one card that can benefit from better alignment methods the  "G" card BPF , band pass filter card. but is 17 trimmers!
Only the Reciever needs a SIG GEN. ever.
The freq. counter in the rig we use WWV to check it. (not a $4200 sig gen)
The old Heath books fail to just tell you what tools you need by simply stating the spec, of the tool, that way, the book of facts last forever , not just in 1975.. and not just in Copenhagen... wow...
Here are 2  examples,: Of what should have been. (
Optional Sig-Gen , covering at least 3.4 to 29.6 (10 to 80) meters, with noise below, -50 to -60db noise below carrier sig. 
Is good enough.  (and you can buy or make a line filter for $15, Low pass, BNC to  BNC, and get better if you desire from just about any SIG  GEN or or DDS) <<< surely all hams can make a filter. no?
For some cheap receivers as this SB104, after all the display decade counter board in this gear is screaming load and clear (conducted and radiated ) noise  with 1mhz and 100khz square waves, unshield.! A hard act to beat ,no?
OPTION meter: 
  • Any old relic VTVM from Heath last made in 1975, why?
  • Any modern a.c VOLT  meter that is 1meg-ohm input Z or better. It need not read in actual dbm at all, but must be able to handle audio signals, in the a.c mode. Most DMM/DVM sold today do ALL THAT. 21st century.
If you have a scope, you can in fact ,set levels to anything you want. No need for a calibrated  $4200 sig Gen at all.

JUMP to My DDS now.

History and things related to DDS. (and VTVM)
The only reason you are here is for the SB11 words, use a
[Radiometer MS-27G ]   to do the "G" card better. 
This is not 1975 , and things change, the low cost solution   is the  DDS chip based board. (that has clean signals, we only need clean ! and guess what ,we heve better filters today, like what AD sells)
There are 3 Heath  books on the SB104,  alignment, ("A" model newer) (the old non "A")  and the Service bulletin, #11 (SB11).  All three radically contradict each other.
The 2 real books, with paper yellow covers do not ask for any , scopes, or SIG-Gens,  only  a near useless(today) VTVM is asked for and is NOT NEEDED at all.  (we use DMMs today in the modern world)
Book 3 is no book really, it's a service sheet.#11
The SB11 document (with comments) calls out a  signal generator sold only in EU, in Copenhagen [Radiometer MS-27G ]   (this SB11 page is an  error, it forgets to say , needed only for repairs, or advanced testing)
Error 2 is not spec'ing out needs, (freqs, and noise and level accuracies, and let the technician decide what to use !) 
Why?  My guess is Schlumberger in the "Kingdom of the Netherlands" , and owned Heathkit starting in 1962. (they are EU oriented)  they also skipped calling out Heathkit SIG-GENS, due they are too poor quality.
The only reason to use an expensive sig gen is due to clean signal and calibrated output, so the lab can do the steps blind, turns the knob and its correct every time.
There are very few good boat anchor sig-gens that still work and fewer yet that  any lab will touch (calibrate it for you ) but this is one.  HP were TOP SHELF THEN and still are.

One only needs a clean RF SINE wave output !,  to do alignment . (levels are easy to set and see , scope it and pad as needed !) If lacking a good scope, the below works.

The HP-8640 sold for $4200 new in 1975 { in $18,000 in today's inflated dollars) and is a good choice, but costs $500 to $1000 used today and most in  need to be repaired then calibrated ($300 more), (broken gears and all)

The HP is a super clean output, but is overkill. (ever look at all the noise in your receiver now? with antenna shorted?)
It uses a tuned cavity to create near perfect sine waves. (you will be hard pressed to find a better SIG-GEN on earth for $1000 and $300 calibration)
HP SPEC: (hard to beat) keep in mind parts inside are not sold now, so we must cannibalize a 2nd gen, to make 1 from 2.

The level range is + 19 to -145dbm  (2volts to 0.013uV) we only need to own cheap pad sets,  to drop our 1v to any where we want.! 
If you own calibrated pads and have a real scope you can infact set RF levels any where you want.
The full magic of the HP 8640 and the men how made it and the cavity drawings, wow. is here.

The  DDS is magical, almost like a modern SDR or DSO scope or DSP , why?
My report  on how to use any  DDS as a SIGNAL GENERATOR or SIG-GEN.
To avoid all that cash  buy a   nice modern scope DSO and a top brand clean  DDS.  No lie I almost had to learn Chinese to buy one.
I also have  a nice, filter chip in a box.  This device in a box with Z matching network and spi linked to say Arduino card.
Today we have at our hands a huge array of choices of filters, from simple LC home made low pass fitlers, to digital chip filters with even internal VGA inside to set gain or attentuation levels
You can filter out any spectrum of noise.
The best filter is called a elliptic reconstruction filter. (but is mostly overkill , many a  DDS  can use this)
 All good DDS have this, it cleans it all up. (but limits upper freq. ) Some 11 element 3 chock Pi filters , mine seen pretty much ok.
In my opinion is better to have a good scope, and do the simple tests/alignments using it for accurate levels. (and to diagnose problems in old gear, nothing beats a scope)
My Scope is so accurate (and auto calibrates) I can use it to check my other instruments to see if they are ok, and not way off accuracy wise.
Buy a used, TEK scope cheap as low as $50-$100 USD, but do not buy tube scopes. even better is  the TEK TDS series.
I recommend a scope 1986 and newer, I do.
This is because they hold CAL, longer, (most are DSO /DSP based and auto calibrated now, saving huge costs in CAL LAB fees)
Keep in mind only the RADIO MDS checks needs super level accuracy ! (1uV spec)
ll  alignment steps are only peaking and dipping signal !  and if wanting better alignment results as the book states, you can find tune the BPF card G, set them and the tricky center points better and more accurately.
This does not take $4000 worth of gear nor even $1000.

Modern Digital multi-meters are up to 10meg  input impedance, now. (some are 100meg , and sold just for that...) 
The newest SB104 service bulletin SB#11 calls out  "a voltmeter with 1meg, input Z or better"  they meant, a.c volts, and Audio rated.
I'd say the Sig gen needs better than, -50db noise below the tuned freqency,  (Heath was silent here , so...)
The volt meter only needs to be in the ball park near 0dbm, -5dbm and -10dbm   it's only looking for peaks and dips and NEVER exact DBm at anytime. (MDS checks excepted)
We only need to use a.c. volts and  at no time do I need a pro grade 600 ohm audio dBm meter, of any like.

By the way, the MSD check is just that ONLY a check its not aligment, it's a check and if wrong, the gear needs repair and that may take the huge list of tools (meters,scopes, siggens listed in SB#11)
The fact is your DMM for $10 on A.C volts will read dBm,  IT WILL , using volts to dBm tables. (seen below , all ways but Sunday)
You only need 3 tools,  DMM 1meg Z or more, A good DDS, and a good scope and a set of pads.
Now consider that the 2 yellow books tell you to use the 100khz, square wave generator (it's built in to the SB104) using the counters wild odd harmonics to align the rig,  tell me how the DDS below can be worse? hark!

First get a  Good scope (mine even does frequencies., and Bevvy of readings as seen here)
The tools needed are: (If you own a good scope we are talking  $50 extra parts, once you own one.)
With this set of tools, I can use my DDS and get a very accurate, 0.5% spec, sign wave frequency with little to no noise. Mine is -50db down, this noise (it is super accurate and will stay that way)
I then set my DDS output to a low value stated in the Heath book:  By using these means.
I pad the output down,.  For the new SB104a book it calls out , -10db (they meant -10dBm)  So I pad down my cheap generator, do the level asked for, using my scope. It's that simple.
What is a pad? the new guy said, well a "Pad" means,  a set of calibrated, inline coax , (I like SMA) attenuators pads, marked on the side for DB and 50 ohms.  Do not use TV parts, at 75ohms. (or telephone parts)
BNC pads for sure are GREAT TOO.
We only need one set of facts to do this with a good scope , what are those facts. see the end of my page on how to know what 0dBm is and -10dbm.  ( the "m", in dBm is 1milliwatt in to 600 ohms)< this is the key to calculating volts.
0dBm is 0.775v., minus -10dbm is 0.245v (rms)  x 1.414 to derive peak voltage.)
 Even if math is hard for you, the tables at the end  of this page solves that magic , as does this next link.
Question of the day?, HOW does one express, zero output in dBm , you can't ,errrr    well sure -1000000 dBm.....  this odd fact, makes some techs go crazy. (heck anyone...)
Question 2: What is 0db  , this is not  a level or a signal, it means no gain, no loss, because dB is only a fraction or ratio.  It's a logrithmic ratio the acts like a humans ears.


This is a 21st Century solution to 1975 problems. DDS is Direct Digital Synthesizer . 
Now the DDS NOISE deal (fear?), I have a line filter, for my DDS,  http://www.minicircuits.com/pdfs/BLP-30+.pdf , as seen here, this makes my signals harmonic free.
We can make low pass filters all day, even tune them to the JOB. (after all ham radios is all about tuning and filters, even an antenna is a complex filter , if you think about that just a minute)
The Receiver Align steps are easy , most steps, just the "G" card is a pain, but if  it works great now, leave it be, as some were factory tuned.  (the clue is no flux seen, on back of "G" card.)
Let me say , I'd not knowingly fiddle those "G" trimmer's, using any 100kHz crystal calibrator, ever. (as the books states)
In most cases doing that; you will make it worse.  For sure trying to tune to harmonic counts at 100khz up to 29mhz,  a whack. good luck with that.(IMO)
The truth is , reading the manual (yellow), Heathkit would   have tuned  up your sent in card for $10, back in 1975,  now those were the golden days.  $10 or $4200 for the SIG-GEN, let me decide that? hummmmmmm

 The manual shows how to find the secret TP test point  to steal the 100kHz out of the Freq counter ,decade counters., easy.  See my slide show and see the "A" card there, and that  test point, I also added that TP to my schematic.
The 104a book (only) shows connecting the external meter to the audio power out driver, this is a very low impedance source and as such never needs any kind of expensive meter to measure this pin, only one that can handle audio, is needed.
The books both tell you  the alignment will be better using a real Signal generator. (what an under statement, IMO)
 The transmit side: 
 The TX alignment is super easy, the VFO is it's own signal generator.  The TX side is truly a breeze , it was on  my SB401, tube Transmitter too.
The RX side needs proper alignment, doing it the book way or use a good Sig gen. (and a voltmeter or scope)

Only the G board is tricky,  with many cards used per (#85-1459-1, #86-1666-1, #85-1708-2) there are 3 generation...  (2 coil per bank, or 3, mine has #1666 , 3 coil banks.)
  In all cases for just the "G" card, the $60 extender card is needed.

My  DDS Photo Review.  I will call it a "GEN" I will call it  "THE CARD"
There are many (too many) AD9850s sold, most of them all have HORRID filters. (why? , I'm not sure!) but most are toy grade.
No lie, any DDS has the potential to make clean RF signals. ( using filters) There are even commerial filters sold. (mini-circuits  has them)
The below works in manual (Manu) with out a PC/Laptop , Using tiny buttons but needs, USB for power. (a power hub, can do that or some USB Cell phone power pack for $3 (a.c line to USB)

I can use any  calibrated scope. (to measure true output of below) 100mhz best scope to own for HF work.  This unit runs on a AtMega16a processor. (much better)
One nice feature on this GEN rig, below is it stores the last setting in EEPROM, so when you power it up, it's where you left it. Nice ! (the 05 seen in the display is my failed attempts to set levels to 05)
Here is mine sold by 2012moon816 on fleabay. ( it comes with a zip file and all source code)
the CARD ! It only needs USB power to work!
This is not a problem the dead Amplitude,, as all good shops have  PADs.  I was told later they have new card, that cures the RED X problem , I wait the baited breath that....
I asked the Chinese to put a real name on the card , like "Golden -Dragon9851v1.0", so we know who made it and what is the model name and what verison.... V1.0) and publish the changes,  
The X is a future feature Dead, the VGA chip to do that is missing. (we know why now)
HOW to use it , in simple steps.?
The USB port connected to Laptop running, a PC command program.
 (the serial drivers must be loaded, and find port # in device manager (mine shows Com7), then set port above to #, then push open radio button here..
Only this Baud works ! Seen above.
The setup process is simple and typical of all such USB devices.
Install  the driver first , then install the appliction seen above, control prog.
You must plug in the USB DDS or you will not see the driver.
Go to Win10 device manager, and see COM port assigned to.
In the example below it's 7, on my laptop it was 4, so we look first then change the port in the command port to # , in this case 7, never 1.
CH340 is the china made usb/serial chip, not the more common USA types. Windows 10 may support it directly , I can't tell this because, my PC has  them ALL installed, tons of widgets here.

If you see and error now at send button pushed, it means wrong port#, or missing driver.(check DM again for correct driver matching this onboard serial chip)
To see comms work just click  send, and see the CARDs serial LED activity light flash. Bingo, comms.(baud must be as seen above correct)
I can sweep or set frequency this way or PCB  physically  push the MAN button, then up button or down, to raise/lower Freq. on the unit itself. No PC needed.
The SINE wave seen on 100mHz scope is really near perfect, no distortion seen at all, like many other CARDs  have. that lack proper filtering.
The Command program software: (7/23/2013 build dates)
The variable levels output feature is inop  so I use real external  PADs. 
The maker responds, 10/7/2016   "the VGA chip failed to perform, so they removed it".  (and forgot to sell it as , VGA feature, removed)

Here is a $10 chip that works perfect.
They sell digital pad chips now too. So that too can be used.  250- 500mV out is plenty to start with, we only need to pad that down. (levels vary by frequency)
The sweep feature works great, and is very useful.  See the actual 3 stage Pi,  filters it uses below and output spectrum below.
I use this only for tuning my BPF on Heath cards, coil/cap-trimmed ,band pass filters. nothing more...  (but MDS can be done too, aka, receiver sensitivity. too , padding correctly !)
One upgrade would be to ADD, a steel box enclosure , a faraday shield and a SMA jack on it , will stop all RFI from leaving said box. 
I'd pay $100 + for  DDS in a faraday box (keeps those nasty clocks inside at bay) and a working VGA chip on board with software to make it work.
Or sell the VGA card seperate with software, we use SMA in and out, and bingo, no more need for PADS.
BTW: the unit defaults to AUTO mode.That means you can run it , without a display, remotely.

Next  is: MDS, for fun I measured it. Doing the classic Minimum Discernible Signal test.  (pad it down until you just can't hear it anymore (threshold point check)
I then use pads as needed, to pad down the DDS SMA pin  signal . (a tedious pile of pads can do this)  but I now have the below device from a wrecke URM25D

The acutal  MDS spec, in your radio manual under specifications.
My SB104 "less than 1uV for 10dB signal plus noise to noise ratio for ssb ops."page 316. as and example.
See my signals below , and a way to pad signals and how to measure DBm on a scope.
Keep in mind I'm only tuning relic coil/cap BPF filters, not designing and tuning crystal lattice filters or the like.
Ex. Military gear,  I had one in 1968 USN. (ship board gear)
2  From a URM25d sig gen, scrapped.  EACH step drops the analog  voltage a factor of 10 times or 20dB.  5x20=100db total in 5 steps.
Step one above is 1:1 ratio, I tried to measure  insertion loss and see ZERO. with my scope. all tests at 4.000mHZ.
Here are my steps to do simple MDS. (low buck way)
DDS connected it J103.   (all readings for bolts are PEAK)
I used my scope at TEE fitting at input, of above and below., I have 180mv  (this is the loading factor on the DDS, yes it's a gutless. but works)
The sets all rules of measurements going forward, The 180mv is solid and does not change, as I step.
I then go to step 2 (20db) (1 click) and get 18mV  (10x less dead on, the URM works)
Step 3, the scope runs out of steam but can see its about 2mv, (I must trust the devices from here on out, do to lack of $5000 HP power meter to certifiy each decade on the URM and tag it, with offset errors)
I connect the SB104 ANT jack, to center BNC below, and find the signal is loud and clear CW mod, at the last step, 5, or 100db. and is 1.8uV, (actual)
I then add a 6db pad (gold screw-on SMD) and this is my limit,  my MDS is 106dB,   (6 dB down, is 1/2 of 1.8uV or 0.9uV) spec is 1uV. I pass the test.  I can just hear the beat tone. at 4.000mHz.
This is an Ears only MDS check. (not doing s/n ratios and all that)
Okay I see someone doing what I am doing on the web. here.
The URM25d (all silver over brass), cable on top is BNC female. input.   (A used Toggle strip atten,below is over $60)

If you do this test, all other dB tests in the book are way more simple this is an acid test, of you, the gear and your tools. 
So when the book asks for -10db audio at the Audio RCA PATCH out jack. You know what that is, it's -10dBm at 600ohms , as that is audio rules. or 0.244 V(rms)
You only need  DMM , rms a.c. volt meter or scope to do that.

Precision is needed and comes in 2 forms,  a very expensive signal generator, with calibrated freq and levels, in dBm.
or a scope and calibrated attenuators.

I must pad it (verb) first  to a known even dbm.  using this link  
So What did all this cost:
DDS  $40
Scope, all good shops have 1 good scope or if not, you get to buy $1000 HP SIG GEN, (old) 8640s? (barrow a good scope? or rent one and be $900 ahead?)
URM-25 atten.  $25
Box of pads, $15 see below or a URM or snap switch pad. (for $25 to $60 respectively)
If you build your own gear, test your gear, repair your gear or align it, the DMM is the first buy ($20) and then a scope., in the long run , this method will save you a fortune.
Btw, that is a real  scope not some useless relic tube scope.

End MDS checks, see more below. on sets and using different test equipment.
A Parting shot: if your rigs receiver seems weak on all antenna's the MDS check is next. or if other radios receive ok, then MDS is next)

Other devices that can work here.(all ATTEN are  calibrated.) You get what you pay for in calibrated devices.
PADS 101: or RF inline Attenuator devices. 
We say pad it down -10dB (a verb, to pad down , using noun pad seen here, and all ways seen hear, pads, pad strips or the URM device above..
Then if  I need more? padding or more precision?, (unlikely ever on this SB104 radio, acutal accuracy here are are overkill for SB104, except the measurement test called MDS.)
I use this step switch attenuator , using BNC to SMA adapters.  $60+ used.
This is the best to own. 101dB max.
I then connect this BNC out to my RADIO, Antenna  and tune the band pass. (or what ever)
End PADS 101:

TABLES OR online electronic calculators are best. Seen HERE and HERE and HERE (nice !)
We only use 1 equation,  that works on any Scientific (TI 30xa), calculator, using 20xlog.  (10xlog is for POWER)
Log(10) must means base 10 Log(b) where b means base and is decimal or what called common logs,  the key on mine is marked LOG. (no fear it's just 1 key)
GdB = 20Ělog10(Vout / Vin)     ( In pure English , avoiding Algebra equations.. , this reads as, Gain (or loss) equals 20 times,  the log(10),  of the quotient ( Vout divided by Vin)  do the vout/vin first then use log10 of that. and last times 20)
A test.  Vin it 10 v, out it 1,   so that is 0.1  I hit my  pocket calc, log key and bingo get -1 (minus 1) and times 20 I get -20 so that means -20db loss.  easy no,just one fancy log key to press.
  1. Press 1  , press divide key then 10 and =  equal key.
  2. then press log key.
  3. last press times X key, then  20 and = Equal  key. bingo, db.
End logs, end math, begin tables.  (warning you can not interpolate logs, well not with simple calcs.)
  THE dB rules in table form. (decibel , 1/10 of  Bel)   The below double Tilde marks are ~2 , 6dB double power point and the 3dB double voltage/amps point.  These are only RATIO's!
The right side is voltage, most folks never measure RF amps so , it's mostly volts. "amplitude"
If you had 1volt  rms to begin and then added a pad of  -10dB, that voltage is now, .316 volts rms or about 316 mV . (or times 1.414 for peak seen on any scope)
One more example say 1volt rms to start, pad to -20dB the voltage is now. 10 times less  or 100mV.  (1 v divided by 10 or 1 volt times 0.1 , gets the same anwer)
Most folks never use the power side unless rich, with a power meter, and power attenutators, costing up to $5000 just to do all this... so I will skip this, and just do, Reciever checks at 1mw or less.
Not to forget mentioning this nice page on DBm
Chart 1: dB  almost zero table math here,   I do my MDS test starting with a 100dB pad, (a 100,000 time less signal)

One can buy this power attenuator for $15 a kit.

Or DIY from scratch your own pads, 

Here are some DDS actual signals and spectrum.
 This section shows how to know what 0dBm is and -10dBm or any other, dBm.  ( remember dB is a ratio !,  only, dBm is an actual measureable level , the "m" means 1mw,  1/1000th of a watt)

Here is a chart that solves DB ratios with no pressing of the log key on your calculator. That chart is for free on wiki and 1000s, of sites.
Below is the noise out off my sweeping DDS. AD9850 based. This device is programmed to be a SIG GEN.  for a fraction of price of any SIG_GEN.
This output floats my boat,  I don't need a high priced, SIG-GEN ever. with accurate sine wave, and clean sine waves, only the levels must be see using any good scope. 
Point 2, why not own a good scope that has vast utility unlike a SIG-GEN on the shelf gathering dust for the next 10 year alignment.  My scope can do FFT too. to see spurs..
If the book says drop the signal 10db that means,using my scope drop it by about 3 times (amplitude drop).
 As you can see the Views  below are volts not power,  0.8 to 0.24 is 3 times less. clear as day.and is rms AC. as the dial shows.
Amplitude means volts or amps, Power is watts.  (and has its own power equations)
These 3 photos came with the CARD as files. "sellers snap shots"  Keep in mind much of the noise on cheap spectrum analyzer  are all inside the analyzer, short the input and see that. (then subtract that from below)

The filter used on this  DDS board, that produces this clean output is below:
The DDS 3 stage Pi filter. If you need more, buy more , or make more filters. (low pass HF).

Here is a German made filter:

To learn how filters work in this manor, read a book on how CLASS C linear amps, avoid the gross harmonics created by this mode.  they use filters. Noise and filters are not new to hams, study Class "C" amps and see why.
Or open the ARRL book and read the huge filters pages.  The have Pi filter chart that is HUGE.  Nice work at ARRL!  Buy the book ever 10 years to see new TECHnology, added.
My scope shows: (sine wave and then the FFT view)

Below that noise on the left is scope created.  As are other spikes. If I short the input , we see them remain, so....

This FFT mode is not the best sharp knife in the drawer,  it's only a freebee added to a read DSO scope. FFT button.

Good DDS Links:

Meter madness:
The Hollow state volt (dbm) meter called VTVM.
Any DMM (DVM) made can do this. (any with A.C volts RMS)
or any real scope (real means calibrated, not some useless $20 1950s scope for TV shops.)
Sure and old 1976 (last year) VTVM works, sure.
or any VOM. 1meg ohm input Z or better (impedance)
All that matters is a.c. volts. (and most work at audio rates, 20 to 20k hZ, most do)
The newer Heath book (104a) tells you to use this RELIC VTVM ( , and as you can see, on 0db scale, that 0db is near 0.8vac (rms) and -10db  is 0.24vac and  your scope can read this as 1.414 times (rms to peak) larger (square root of 2 )
That really is all you need to know, just the a.c rms voltage, the peak voltage  using a scope tells you all that.
The scope is better, if the signal is a MESS, full of noise , hum or other gross anomalies, then you instantly go,  gee this is bad and  I must repair this gear now,  not alignment on NOISE, yah!
The schematic on the meter is cave man simple a  1820 year  relic D'Arsonval meter, and a TUBE amp buffer added,  (we do have FET meters, between tube days and FET OP AMP DAYS)
The only claim to fame back in the dark ages, (1960s to 1976 , Ham era) Is the Vacuum tube has 10meg ohm real input impedance something no VOM had then.
VOM  were even more cude, a resistor and the current sucking  D'Arsonval meter, (the meter is a coil, the  meter is just a DC motor, lacking a commutator and sucks lots of current , naked)

As you can clearly see, the meter has no idea what dBm is, only the paint on the dial tells you its 0.8v for 0dbm. and is a standard that. <<< this is my key point here, just print out the dial and use any modern meter.
History:(1 line):
In case you are curious , 1dbm loss is 1 mile of telephone signal loss that humans can just detect by ear. "The scale is for human ears , from 1924 to now. (or light to your eyes same deal) <shortest history of dB.
Note how the dial clearly states  dB  at 600ohms , and 0 means 1mW ! (many meters like this fail to tell you that, on the dial or its matching book)
  They assumed then, that you knew that Decibels is really dBm, but in this case,  is clearly stated on the dial  , but not on all Heath meters.  (this  meter manual also states these dBm 600ohm ,facts)

So now you can wean your self off of VTVM and  no more, use hollow state meters needed ever again. (we use FET METERS  today)
Any modern Solid state Tek scope up to today can measure dBm , and is child's play, just some easy math.
The true levels , accurate are here.
 These are the 2 settings in the 2 book from Heath. (keep in mind that both numbers are BALL PARKED)
Even worse is book 3, or the SB11 Service bulletin, best is to avoid it. (see my comments in this pdf here) 
Now the accurate numbers for dBm, using the logrithm calculators.  So a scope or meter can work right.
0dBm @ 50ohms is : (as the alignment manual demands)
0.224vrms  (times 1.414 to get peak for scopes)   this is not it, its for RF only. I did it both ways just to demonstate how easy it is to get wrong using the wrong impedance.
It is these values for audio tests:
0.775vrms for 600 ohms, see link below to do this easy. The dial shows near .8v
The meter above is only for 600ohms only. (never RF)
-10 dBm padded down  from .775v-rms is 0.244 (see the dial agree there?)
-5dBm would be 0.43558v rms,  the dial  above shows. .44? sure.
Not only that  but,  .8v and .5 volts would work 100% ok in those book alignment steps, it's only shooting from the hip, really.
Go here and
Type in 0 db, then 600ohms, and hit the calculate button.(what's good here is you pick the line impedance your self!)

Point 2,
you could just use this old  dial photo above, to see what 0 means?, see my red tick marks .
 If you look , you clearly see that 0 = about  0.8v. and -10  about 0.24v.  (this is RMS volts) peak is 1.414 times that as seen on all scopes. (this is all ball park readings, do not go OCD here !)
If you have a modern scope , it will even do RMS for you, in fact, has cursors that you can twiddle knobs and then it reads out in peak and RMS.  All for free, no need to punch in  volts times 1.414 on your calculator.
My scope reads frequency the same way, and is  more that accurate enough.
My modern DMM meter on AC volts , I press, the mode and it toggles to DBm , magic..no?( "J.  Fluke" engineering and meter with same name)

Keep in mind dBm has many standards,  50,75, 300, and 600 ohm line impdance standards,  our yellow HEATH book uses 600, telephone mode. 
( each technology uses there own standards, audio, Radio, Telephones,CCTV, Video, FIBER cable it's
acutruments ,etc and more...)
They even use dbm on DC powersupply noise, but at what impedance rules, 50?,  some linear amps are 0.14 ohms dynamically, so how can dBm have meaning there, well it can't.
(they use the more complex forms
| dB = 45 dBuV/m) but can be converted, to voltage.
Later in the newest Heath book, they switch to 50 ohm mode and never once tell you!. (at MDS checks) but is implied in the context, being it's now RF checks and not Audio this time.
The pages on Alignment at the words, PATCH OUT:
Connect any audio rated , a.c voltmeter scope or VTVM, or FET based DMM.   (in my case the jack connected is audio amp out) BTW this is a LOW Z circuit ,  so, a VTVM was overkill.
Then set 0dbm at about 0.8vac, rms,  (▒20%)  that means its a ball park reading., see? and is NOT STATED clearly ,in the books, sadly.
What is not said is the signal is set crudely, so that signal is more than noise, and not so large that you saturate(overload) some radio internal amplifiers.
At no  time are you setting exact  levels at some exact dBM levels, you are only peaking or dipping coils (trimmers). 
All you need  here, IS just that  dial with red tick marks, once you know that and seen that, you are good to go.!
End dB.

Alternate ways, and links. (really are the best of the best, IMO)

The great K5BCQ alignment all 3 ways, scope, sig.gen, swept , spectrum analyzer by,  KEES A TALEN

Thanks Kees ! I love your simple and clear, easy to read web site, even very efficient Black/white photos, of the scope screens , nice.

Then there is JANIS nice site,  and a great review on the SB104,  priceless advice. THANKS!

All FILES and documents, related to the above .

The SB104 books are here, on alignments.(both books)



Here is my scope fix page.( if you can't afford a scope buy a broken one and fix it,  buy TEK)

End . DDS ,magic

YMMV, your meter may vary,  !
the guy with the rope is Mr. Resistance. (or Impedance) (for water,  volts is PSI (pressure),  amps is gallons per minute and ohms is a orifice restriction.)
When this orifice goes to zere diameter, that is an INSULATOR. 

Keep in mind , if it were not for Mr. Resistance,  everything would happen all at once,  even explode. (lightening is a prime example of that)

version 2  9-27-2016    , any questions or need help, contact me on my EHAM.com , page, under my K5JXH,  name, there. I will respond.  (50+ years as an electronics tech.)