EEVblog #306 - Jim Williams Pulse Generator

[carpelux]

Thanks a lot!

[jpb]

I received my kits yesterday.  Thank you! 

I built one up this morning and put it on my fastest [working] scope, a Tektronix MSO3034.   Photo attached.

I'm looking forward to getting my kit. It seems to produce fairly accurate bandwidth measurements on scopes in the 100MHz to 500MHz range.

Your measurements are for a rise of 7 times. The tau of the pulse generator is around 200psecs for 20% to 80% or 4 times so for 7 times I think this would increase to (ln(7)/ln(4))*200psecs or around 280 psecs so the net rise from your measurements is around sqrt(980^2-280^2)psecs or 940 psecs.

The Bandwidth is then approximately (ln(7)/2.pi)/940psecs = 0.31/940psecs = 330MHz which is 10% up on the nominal 300MHz.

All very believable.

[EV]

  Photo attached.

Risetimes marked. 

[John_L]

Thanks Vince

[jpb]

Is it ok to connect the pulser straight to the 50ohm input of an oscilloscope rated at 5Vrms (for 50 ohms)?

I have a WaveJet and the specs don't list a peak voltage for 50 ohms, just the rms value of 5V and I guess that though the peak voltage for the pulser is 10V it only lasts 5nsecs and the duty cycle is low. (What is the duty cycle?)

I haven't got my kit yet but I don't possess any attenuators at present and was wondering whether to get a set (they are quite pricey).

[EV]

Is it ok to connect the pulser straight to the 50ohm input of an oscilloscope rated at 5Vrms (for 50 ohms)?

I think that there is no problem. My Tektronix TDS3032 has max 5 VRMS with peaks < ±30 V. Problem was that the max vertical division is 1V/div. That is why the pulse can not be seen completely without attenautor. With attenautor the uppper side of the pulse is not horisontal. Look at my earlier post:

https://www.eevblog.com/forum/blog/eevblog-306-jim-williams-pulse-generator/300/

[free_electron]

It depends on the scope ... It will definitely clip the signal.
I had the same problem when doing the measurement. I used a real divide by two attenuator during the test.

The attenuator is 50 ohms both ways. Meaning it is 50 ohms towards the pulse generator and also matches the 50 ohms of the scope. (Plus we 'calibrated' the attenuator away.)

[EV]

I have got now 6 dB attenuator. It is much better than that 20 dB attenuator which I used earlier. The upper side of the pulse is much nearer horisontal line now. I have ordered also 3 dB attenuator which is probably still better and with it the whole pulse can also be seen.

[jpb]

It depends on the scope ... It will definitely clip the signal.
I had the same problem when doing the measurement. I used a real divide by two attenuator during the test.

The attenuator is 50 ohms both ways. Meaning it is 50 ohms towards the pulse generator and also matches the 50 ohms of the scope. (Plus we 'calibrated' the attenuator away.)

I guess I'll go with an attenuator to be safe.

The WaveJet that I have states that 5Vrms is a maximum for the 50ohm input but it goes up to 2V/div on 50ohms so presumably can show 16Vp-p so should be able to show a 10V pulse without clipping (I'm assuming that you can apply an offset of around -5V.)

SMA attenuators appear to be much higher frequency than BNC ones (18GHz vs 1 or 2GHz) for the same sort of price but I guess it makes little difference as the SMA to BNC adapter is probably only good to a GHz and my scope is 350MHz anyway.

(Hameg do a reasonably priced set of 4 attenuators and a 50ohm pass through rated at 1W and  1GHz with BNC connectors designed for their scopes.)

[free_electron]

Anyone not received his pulsegen yet ?

(Apart from the last 8 buyers. Those leave today or tomorrow. I packcaged them over the weekend)

[fmaimon]

As I'm on vacations, I won't know for another 2 weeks, but I wouldn't be surprised if i still havn't arrived, as the Brazilian Post and customs are very slow... As soon as I get home I'll check.

[w2aew]

OK, here are my results from pulse tester #6 (thanks Scotto!).  First is the result using my old Tektronix 2467 which is a 350MHz analog scope.  Cursors are on the approximate 10-90% points.  The next two images are using a homebrew 20dB attenuator (which may slightly slow the results), on a 1GHz Tektronix MDO4104-6 and on a Tektronix 8GHz MSO70804C.

[hgg]

free_electron, hi !

Can I have a calibrated one as well ??  Do you sell them assembled?
What is the total cost to Greece?

[SeanB]

W2aew, I think the scope is the limiting factor here...........

Hgg, I think Vince does have some left, you will have to PM him to see.

[jpb]

W2aew, I think the scope is the limiting factor here...........

That is true for the first two, but not for the 8GHz scope.

Vince's measurements gave rise times for 20% to 80% whilst W2aew's are for 10% to 90% so will be (ln(9)/ln(4)) times as long approximately (at least that would be the case for a simple exponential type rise as with a Gaussian filter - I'm not so sure about the case for maximally flat input responses as in higher frequency scopes).

So 234psecs for 10% to 90% would correspond to only around 148psecs for 20% to 80%.

[EV]

W2aew, fine scopes and pictures! 

[hgg]

Thanks SeanB, I will.

[BravoV]

I will post the final cal screenshots this weekend. Last batch goes out the door tomorrow. Fmaimon was in the last batch.(last week)

I'll post an updat tonight. Everything os put the door. There are kits and cals left and a few bare boards as welll. How many ... I need to cou t em.

Ah yes. I will post the other screenshots today.

Shameless bump. 

[free_electron]

I need one of three things :
- invent time travel
- clone myself so there's two of me
- days of 48 hours ...

I'll try posting the rest tonoght.

[AlfBaz]

I need one of three things :
- invent time travel
- clone myself so there's two of me
- days of 48 hours ...

I'll try posting the rest tonoght.

Gees you're a wimp, there's at least another 8 hours you are wasting with sleep!

[Teneyes]

Be Nice

Be Positive to catch more Free Electrons

[free_electron]

Allright, here's all the cal certificates (shees,h i'm missing out some quality time floating in the pool...)

Serial number 1
Serial number 2
Serial number 3
Serial number 4
Serial number 5
Serial number 6
Serial number 7
Serial number 8
Serial number 9
Serial number 10
Serial number 11
Serial number 12
Serial number 13
Serial number 14
Serial number 15
Serial number 16
Serial number 17
Serial number 18
Serial number 19
Serial number 20
Serial number 21
Serial number 22
Serial number 23
Serial number 24
Serial number 25
Serial number 26
Serial number 27
Serial number 28
Serial number 29
Serial number 30
Serial number 31
Serial number 32
Serial number 33
Serial number 34
Serial number 35
Serial number 36
Serial number 37
Serial number 38
Serial number 39
Serial number 40
Serial number 41
Serial number 42
Serial number 43
Serial number 44
Serial number 45
Serial number 46
Serial number 47
Serial number 48
Serial number 49
Serial number 50
Serial number 51
Serial number 52
Serial number 53
Serial number 54
Serial number 55
Serial number 56
Serial number 57
Serial number 58
Serial number 59
Serial number 60
Serial number 61
Serial number 62
Serial number 63
Serial number 64
Serial number 65
Serial number 66
Serial number 67
Serial number 68
Serial number 69
Serial number 70

[c4757p]

(Apart from the last 8 buyers. Those leave today or tomorrow. I packcaged them over the weekend)

Have you mailed these yet?

[Teneyes]

@BravoV
  Will you be nitpicking about the Standard deviation in the values of Risetimes of Free's workmanship

[JimmyMz]

Free_Electron, you have a PM.