Tek tds 3054b vs tds 744a

[cncjerry]

I bought an NI PCI-GPIB adapter card from a guy in Sweden.  It as $50USD shipped and works perfectly.  He had 4 left a few weeks ago, I am buying another one today so you better hurry...

I am opening up  my 744a as I type, going to try to the mod.

Thanks again.

[MarkL]

..
1) on the 3054b, I read tektronix released/unlocked all the module software in version FV4.1.  I have 4.1 on my scope and don't see the advanced analytical functions under the math menu.  Were there two versions of 4.1 or was that not included?
..

I do not know about 3054b modell but in TDS3032 only Adv trigger and FFT modules are unlocked with the latest FW, nothing else.

That's interesting your scope is running 4.1.  According to Tek's web site, the latest firmware for both the TDS3000 and TDS3000B is 3.41.  Is 4.1 available for download somewhere?

Even for the TDS3000C (latest is v4.26), Tek is still selling the same Math, Limit, Telecom, and Video keys so I assume those have not been unlocked in any version of the firmware.

[cncjerry]

Yes, I mistyped the version number, it is 3.41, not 4.1 as stated.  Tek calls this FV3.41. 

I took out the resistor as the previous posted suggested in the 744a.  It now presents as a 784a and moves smoothly into 4Gs/s sampling.  Pretty cool.

This doesn't make sense, though and I wonder if it is calibration related:   The risetime on the probe calibration signal now reports as 1.183 after the change where it was 1.340 before. All the settings are the same with the exception of the 4Gs/s mod.  I have repetitive signal turned on and the ET is reporting as 250Gs/s.  I don't remember it being that high in the past.

Another question:   On the 744a there is what looks like a thin PC board sensor around the BNC jacks.  It looks like the probe pin wore the surface coating off a little.   Is this considered a problem?    I've seen those PC board/cables for sale, and thought they were a little expensive.

[nctnico]

Hmm. I'm starting to regret selling my TDS744A even more  . TiN has told me about the 1GHz hack before but I never got round to try and more or less forgot all about it.

[David Hess]

This doesn't make sense, though and I wonder if it is calibration related:   The risetime on the probe calibration signal now reports as 1.183 after the change where it was 1.340 before. All the settings are the same with the exception of the 4Gs/s mod.  I have repetitive signal turned on and the ET is reporting as 250Gs/s.  I don't remember it being that high in the past.

I doubt this could be related to any increase in sample rate.  The original transient response calibration of the oscilloscope was probably only done up to 500 MHz so when configured to 1 GHz, it results in a peaked response which will make the transition time look faster than it really is.  You could test for this using a leveled signal generator or reference flat top pulse generator.

[nctnico]

AFAIK there is a programmable bandwidth filter in the input stage which is disabled by the hack. If you look in the TDS520B component service manual you'll see the input stage has several bandwidth setting pins.

[tinhead]

one of the calibration steps is HFCal, which is unique for 500MHz ang 1GHz. Check the fied adj software, and all these *.con files and you will see all the differences during clibration for different models. So yeah, sampling as said before is getting increased, but one can have bigger skew between channels and of course bw is not given. Only calibration can change this. When there are errors due unmatched input hybrids during calibraion, one can comment out hfcal in field adj software and run it again. I don't have any TDS700x anymore, so can't help more that this what i already did in the past.

Attached cal data from perfect working TDS754A. It would be still great if someone with TDS784A could dump the cal data.

[cncjerry]

Hmm. I'm starting to regret selling my TDS744A even more  . TiN has told me about the 1GHz hack before but I never got round to try and more or less forgot all about it.

Nicotine,  Yes, let me tell you, most people that come in my lab look around and their eyes settle on one of two things: my 744a; or, my perfect, and I mean perfect, HP-4815a RF Vector Impedance Meter with probe, original cardboard Impedance calculator and manual, I might add. It would be a toss-up between my first born and the 4815a if I had to catch one or the other.  Not to get to far off track,  but I could power on 4815 and measure multiple circuit nodes before I find the cables for my  Advantest r3131a and tracking generator.

And since I am rambling as usual, the other thing I get a lot of comments on is my retro Dekabox.  The picture doesn't do it justice as I cleaned it since.

I'm was going to bid on a TDS5034a that just went for $1,017 on ebay.  It has an error though and I have a stack of errors waiting to be fixed already.  I like that scope.  Another one went for about $1000 a few months ago and all it needed was the software  CD (that Tek wanted  $350 for, by the way).

[MadTux]

Anyone here has the schematic for the attenuators of the TDS 744A/754A/784A?

I'm now a happy owner of a 744A that became a 784A by removing the R1061 resistor from the back of the acquisition board.

Works great apart from the damaged LCD shutter and a problem with the CH1 attenuator. The guy on ebay said something about purple spots on the screen and I was unfamiliar with the LCD shutter technology and therefore thought that somebody was playing with magnets or spilled some ink as I believed that Tek used a normal color CRT.
But I can live with that, since the CRT itself is still in perfect condition and a few purple spots don't really bother me in seeing the traces on screen. If the problem gets worse, I'll simply cut the shutter off and use it in monochrome white mode. Not really nice, but usable.

The other problem I'd like to fix is the broken attenuator of CH1. The relays are still clicking, but I loose the signal if I switch between 100mV/div and 200mV/div and only see small capacitive spikes in 200mV/div mode.
This makes me think that those numerous gold pins in the plastic connector, the amp, ADC ... of CH1 is fine and one of those 4 yellow relays has bad contacts. Switching between AC/DC doesn't change it.

Since there's a black chip on the ceramic assembly and I'm unsure if it has anything do do with the relays, I don't want to use my usual method of cleaning relays by  applying 20V@100Hz _|¯|_|¯|_ from a function generator for a few minutes to a few h until the oxides on the relay have vibrated off.
Maybe someone has the schematic, so I can see which relay I need to torture or unsolder

[cncjerry]

There's a guy on the tek scopes forum that replaced his 744a display with a standard LED type and it looks and works great.  He listed the parts and it was like $150. There are some guys over there in that forum that know that scope inside and out and could help with the attenuator problem.

Jerry

[curiosu]

Dear Tek wizards, I've got a TDS 724A, is there any hack/mod that will increase the sample rate ?

 Many thanks, MC

[dnmeeks]

you can even get more on tds700A series, with 4 resistors (in your case from 744A to 784A only one resistor on acq board) one can get 4GSa/s for single channel and 2GSa/s when 1or 2 and 3 or 4 ch enabled. The bandwidth increase to 1GHz must not work always.

I love this! I have a TDS754C, and wanted to try this mod. I didn't see any jumper settings (R1061-1064) for that particular model / hack, so I mapped all of the settings, just in case anyone else needs them:

R1061 R1062 R1063 R1064 is the order.
1 = INSTALLED
0 = Not Installed
The text (ie "TDSxxx") is the model displayed on the CRT after booting. TDSxxx is literally what some of them said, so I assume this is either a test mode or just not supported / unpredictable.

0000: TDSxxx
0001: TDS784C
0010: TDSxxx
0011: TDSxxx
0100: TDSxxx
0101: TDS724C
0110: TDS794
0111: TDS794
1000: TDS754C
1001: TDS754C
1010: TDS724C
1011: TDSxxx
1100: TDSxxx
1101: TDSxxx
1110: TDS754C
1111: TDS754C

In my case I left them at 0001 for TDS784C.
I verified rise times: before: 480psec, after: 270psec!!! That equates to >1GHz analog bandwidth!
I am going off to figure out how to calibrate this thing now.
Great hack, I can't believe it was there the whole time.
Dan

[eKretz]

I know that this is a very old topic, but there is missing relevant information regarding upgrading the TDS scopes to higher bandwidth versions in case anyone runs across it. The ID jumpers/resistors are NOT the only thing that needs to be done in order to upgrade the scopes. There are coupling capacitors in the signal path also that need to be removed to achieve the higher bandwidth. These are listed in another more recent post which can be readily found if you do a search here at EEVB. Several posters in this thread have apparently performed only half of the work, (the jumper swaps) which is why they see no significant improvement in risetime numbers. After the coupling caps are removed also and the calibration is performed the scopes will indeed achieve 1GHz B/W.